Should both toxin A and toxin B be detected for better sensitivity in diagnosing Clostridioides difficile infection?

Should Both Toxin A and Toxin B Be Detected for Better Sensitivity?

Yes, detecting both toxin A and toxin B improves diagnostic sensitivity compared to detecting toxin A alone, but even combined toxin A/B detection has suboptimal sensitivity (52-82%) and should not be used as a standalone test. 1

Key Performance Differences Between Toxin Detection Methods

Toxin A/B Combined Detection vs. Toxin A Alone

• Well-type EIAs detecting both toxins A and B demonstrate mean sensitivity of 82% (95% CI: 0.79-0.84) and specificity of 97% (95% CI: 0.97-0.98) when compared to cytotoxicity assay 1
• Membrane-type EIAs detecting both toxins A and B show lower mean sensitivity of 72% (95% CI: 0.69-0.74) but maintain high specificity of 98% (95% CI: 0.97-0.98) 1
• Toxin A-only detection assays (like Clearview tox A and Triage tox A) demonstrate even lower sensitivity of 61-82%, missing a substantial proportion of true CDI cases 1

Clinical Significance of Detecting Both Toxins

• Both toxins A and B are primary virulence factors that cause disease through glucosyltransferase activity, with toxin B now recognized as essential for virulence 2
• Only toxigenic strains producing these toxins cause disease—non-toxigenic C. difficile strains are non-pathogenic 2
• Some C. difficile strains may produce predominantly toxin B with minimal toxin A, making toxin A-only assays particularly problematic 2

Critical Limitation: Even Combined Toxin A/B Detection Is Insufficient

Unacceptably Low Sensitivity as Standalone Test

• When compared to toxigenic culture as the reference standard, well-type EIA toxin A/B sensitivity drops to only 66% (95% CI: 0.61-0.71), and membrane-type EIA toxin A/B falls to 52% (95% CI: 0.47-0.57) 1
• This means that nearly half of true CDI cases would be missed using membrane-type toxin A/B assays alone 1
• Recent SIMOA technology toxin assays show improved sensitivity of 76-77% compared to conventional EIA's 48%, but this still misses approximately one-quarter of cases 3

Poor Positive Predictive Value in Endemic Settings

• At endemic CDI prevalence of 5-10%, most toxin A/B assays have unacceptably low positive predictive values ranging from 28-77% 1
• Only the most specific tests (XpecT A/B with PPV 1.00 and Tox A/B Quik Chek with PPV 1.00) achieve acceptable PPV at low prevalence, but these have limited validation data 1
• At 5% prevalence, Premier tox A/B shows PPV of only 55%, meaning nearly half of positive results would be false positives 1

Recommended Two-Step Diagnostic Algorithm

Step 1: High-Sensitivity Screening

• Screen first with either GDH enzyme immunoassay (sensitivity 90-94%, specificity 87-90%) or NAAT/PCR (sensitivity 92%, specificity 96%) 1, 2, 4
• GDH detects all C. difficile strains (both toxigenic and non-toxigenic), making it an excellent screening tool with high negative predictive value 2, 4
• A negative GDH or NAAT result effectively excludes CDI with NPV of 98-99% 4, 5

Step 2: Confirmatory Toxin Detection

• For positive screening results, confirm with toxin A/B detection to distinguish active infection from asymptomatic colonization 1, 4
• This confirmatory step addresses NAAT's limitation of detecting asymptomatic colonization (specificity for CDI only 95%) 3
• The two-step approach achieves optimal diagnostic performance with sensitivity of 91% and specificity of 98% 5

Performance of Two-Step Algorithm

• Using Triage GDH (sensitivity 0.94, specificity 0.87) followed by Premier Tox A/B (sensitivity 0.92, specificity 0.96) in a population of 10,000 with 5% prevalence results in only 1% overall error rate 1
• This yields 432 true positives, 9,451 true negatives, 68 false negatives, and 49 false positives 1
• The combined algorithm achieves PPV of 82-85% and NPV of 98-99% at endemic prevalence rates 4, 6

What If Only One Toxin Is Detected?

Clinical Interpretation

• If only toxin A is detected without toxin B testing, you risk missing toxin B-predominant strains, which can represent a significant proportion of CDI cases 2
• If only toxin B is detected, this is clinically sufficient since toxin B is now recognized as essential for virulence and is detected by the reference standard cell cytotoxicity assay 2, 4

Practical Recommendation

• Always use assays that detect both toxins A and B rather than toxin A alone 1
• If your laboratory only has toxin A detection available, recognize this has sensitivity of only 61-82% and implement the two-step algorithm with GDH or NAAT screening 1
• Never rely on toxin detection alone as a standalone test regardless of whether one or both toxins are detected 1, 4

Common Pitfalls to Avoid

• Do not test asymptomatic patients or those with formed stools, as this detects colonization rather than infection 5
• Do not perform "test of cure" after treatment, as patients may shed spores for up to 6 weeks after successful treatment 5
• Do not interpret positive NAAT alone as definitive CDI without clinical correlation, as specificity for active disease is only 95% due to asymptomatic colonization 3
• Do not use toxin EIA as a standalone test in endemic settings (5-10% prevalence) due to poor positive predictive value 1