ALT-triggered reflex AST testing in health check-ups: A decision curve analysis

Area of Science:

• Clinical biochemistry
• Health screening policies
• Diagnostic yield optimization

Background:

• Routine health check-ups often include co-ordered aspartate aminotransferase (AST) and alanine aminotransferase (ALT) tests.
• Universal AST testing may yield limited clinical value in asymptomatic individuals, prompting evaluation of alternative strategies.

Purpose of the Study:

• To evaluate the clinical utility of ALT-triggered reflex AST testing using decision curve analysis (DCA).
• To identify implementable screening policies that optimize AST testing during routine health check-ups.

Main Methods:

• Retrospective analysis of 46,059 adult health check-up records with paired ALT and AST results.
• Simulation of reflex AST testing strategies based on predefined ALT thresholds (15-60 U/L).
• Assessment of clinical utility via DCA, defining pathological AST using sex-specific upper reference limits (URL).

Main Results:

• Sex-specific ALT triggers (≥19 U/L for females, ≥26 U/L for males) at a threshold probability of 1% reduced AST testing by 53.3% and 37.3%, respectively.
• These policies maintained high sensitivities (97.9% for females, 98.0% for males) with minimal missed pathological AST cases (11.7-14.9 per 10,000).
• Reflex strategies demonstrated greater net benefits than universal testing across various threshold probabilities.

Conclusions:

• ALT-triggered reflex AST testing effectively reduces low-yield AST measurements in health check-ups.
• This strategy achieves substantial reductions in testing with minimal impact on screening sensitivity.
• DCA offers a transparent method for selecting sex-specific, screening-focused policy thresholds for AST testing.