Streamlining cardiopulmonary exercise testing for use as a screening and tracking tool in primary care

Area of Science:

• Cardiopulmonary physiology
• Medical diagnostics
• Algorithm development

Background:

• Cardiopulmonary exercise testing (CPET) is valuable for assessing disease severity but limited by specialist availability and data interpretation challenges.
• Accurate patient classification is crucial for effective clinical management of cardiopulmonary diseases.

Purpose of the Study:

• To evaluate an automated disease likelihood scoring algorithm system.
• To assess the algorithm's performance using a simplified step-test protocol for diagnosing heart failure (HF), pulmonary hypertension (PAH), obstructive lung disease (OLD), and restrictive lung disease (RLD).

Main Methods:

• Developed and tested a novel algorithm for disease likelihood scoring.
• Collected breath-by-breath ventilation, gas exchange, oxygen saturation, and heart rate data during submaximal step-testing.
• Compared algorithm-generated patient scores against expert clinician evaluations for HF, PAH, OLD, and RLD cohorts.

Main Results:

• The algorithm demonstrated strong correlations with expert clinician assessments for all tested conditions: HF (r=0.89), PAH (r=0.88), OLD (r=0.70), and RLD (r=0.88) (all P<0.01).
• The algorithm successfully differentiated major disease pathologies.
• The automated scoring system showed high concordance with clinical specialist diagnoses.

Conclusions:

• A simplified, automated disease algorithm scoring system utilizing step-testing is a clinically relevant tool.
• This algorithm closely correlates with expert clinician assessments for identifying the likelihood of HF, PAH, OLD, or RLD.
• The findings suggest potential for wider adoption of automated CPET analysis in clinical practice.