Can the Candida score still be used to guide initiation of antifungal therapy in critically ill patients?

Can the Candida Score Still Be Used to Guide Antifungal Therapy in Critically Ill Patients?

The Candida score has significant limitations and should not be used as the sole criterion for initiating antifungal therapy in critically ill patients, particularly in COVID-19 patients where it may not have a role at all. 1 While the score can help stratify risk, empirical antifungal therapy decisions should be based on a broader clinical assessment including multiple risk factors, surrogate markers, and clinical presentation rather than relying on the Candida score alone. 1

Understanding the Candida Score's Original Design and Limitations

The Candida score was developed in 2006 and validated in 2009, incorporating four predictors: surgery, multifocal colonization, total parenteral nutrition, and severe sepsis (2 points for clinical sepsis, 1 point for each other factor). 1 A cutoff value of 2.5 demonstrated sensitivity of 81% and specificity of 74%. 1

However, the score has critical limitations that restrict its clinical utility:

• The score demonstrates high specificity but low sensitivity, meaning it may miss patients who actually have invasive candidiasis. 1
• In COVID-19 patients specifically, there was no difference in Candida scores between those with and without candidemia, and the score may not have a role for early detection in this population. 1
• Over one-third of patients with candidemia died before receiving antifungal therapy, with 28-day mortality significantly higher in COVID-19 patients (87.5% vs 67.9%). 1

When to Initiate Empirical Antifungal Therapy

Empirical antifungal therapy should be started immediately in critically ill patients with septic shock and risk factors for invasive candidiasis, regardless of Candida score. 1, 2 Mortality approaches 100% in septic shock patients without adequate source control or antifungal therapy within 24 hours. 1, 3

Key Risk Factors to Assess (Beyond Candida Score):

• Multifocal Candida colonization at multiple body sites 1, 2, 3
• Recent broad-spectrum antibiotic exposure 2, 3
• Central venous catheters (indwelling) 1, 2, 3
• Total parenteral nutrition 1, 2
• Recent major surgery, particularly abdominal 1, 2
• Necrotizing pancreatitis 2, 3
• Prolonged ICU stay 1, 2
• Corticosteroid use 1, 3
• Immunocompromised status (neutropenia, transplant, diabetes) 2
• Dialysis requirement 3

Recommended Approach for Empirical Therapy Decisions

Use a comprehensive clinical assessment rather than Candida score alone:

1. Assess for septic shock with risk factors: If present, start echinocandins immediately (anidulafungin 200 mg loading then 100 mg daily, micafungin 100 mg daily, or caspofungin 70 mg loading then 50 mg daily). 1, 2, 3

2. Evaluate colonization status: Multifocal Candida colonization combined with clinical risk factors warrants empirical therapy. 1, 3

3. Consider biomarkers when available: Beta-D-glucan has high negative predictive value and can help guide therapy decisions. 1 Combined mannan antigen and anti-mannan antibody testing shows 83% sensitivity and 86% specificity. 1

4. Integrate Candida score as one component: A score >3 suggests higher risk (17.6% invasive candidiasis with score=4,50% with score=5), but scores ≤3 do not exclude infection. 4

First-Line Empirical Therapy Selection

Echinocandins are the preferred empiric therapy for critically ill ICU patients with suspected invasive candidiasis. 1, 2, 3

• For critically ill/septic shock patients: Echinocandins are mandatory first-line (strong recommendation, moderate-quality evidence). 1, 2
• For hemodynamically stable patients without recent azole exposure: Fluconazole 800 mg loading then 400 mg daily is acceptable. 1
• For patients with azole-resistant species risk or recent azole exposure: Echinocandins only. 1, 2

Duration and Discontinuation Criteria

Continue empirical therapy for 2 weeks in patients who improve clinically. 1

Discontinue empirical antifungal therapy at 4-5 days if:

• No clinical response to therapy 1
• No subsequent evidence of invasive candidiasis 1
• Negative non-culture-based diagnostic assay with high negative predictive value 1

Critical Pitfalls to Avoid

• Never delay antifungal therapy in septic shock waiting for Candida score calculation or culture results. 1, 2, 3 Prompt therapy within 24 hours is associated with 50% reduction in mortality. 1
• Do not rely solely on Candida score in COVID-19 patients, as it has not been validated in this population. 1
• Do not use Candida score as the only criterion for withholding therapy in patients with clinical sepsis and multiple risk factors. 1
• Always remove central venous catheters in nonneutropenic patients with candidemia—this is mandatory, not optional. 1, 3

Practical Algorithm for Clinical Use

Step 1: Patient has unexplained fever/sepsis despite antibiotics in ICU setting

Step 2: Assess for septic shock + any risk factors → YES: Start echinocandin immediately 1, 2, 3

Step 3: If no shock, assess risk factor burden (≥3 risk factors including multifocal colonization) → YES: Start empirical echinocandin 1, 3

Step 4: If 1-2 risk factors only, calculate Candida score and obtain biomarkers if available 1

• Score >3 + clinical concern → Consider empirical therapy 4
• Score ≤3 + negative biomarkers → Monitor closely, hold empirical therapy 4

Step 5: Reassess at 4-5 days and discontinue if no evidence of infection 1