Network Analysis For Diagnosis And Risk Stratification Of Cardiopulmonary Patients With Exercise Dysfunction

Exercise intolerance is a heterogeneous condition associated with increased morbidity and mortality across the spectrum of cardiopulmonary disease. Invasive cardiopulmonary exercise testing (iCPET) measures gas exchange, hemodynamics, and metabolic data to define exercise physiology; however, overlapping clinical diagnoses in iCPET are common. We tested the hypothesis that network modeling of iCPET data in silico identifies distinct exercise pathophenotypes in patients. Data from 757 individual iCPETs (2003-2015; 44% male, age 57±5 yrs) including 73 clinical and exercise variables were analyzed retrospectively. From an exercise network of 39 variables and 98 correlations (