Circulating Proteins as Predictors of Improvement in Physical Function in Hospitalized Older Adults with Geriatric Syndromes: Findings from the REHAB-HF Trial
Damluji, A. A.; Bruce, S. A.; deFilippi, C. R.
Show abstract
IntroductionCirculating proteins may augment prognostic scores and provide mechanistic insights for cardiovascular disease. Their role for adding precision to predicting response to therapeutic exercise training is less certain. Older adults with HF frequently experience geriatric syndromes, which may influence response to a rehabilitation intervention (RI). We sought to determine if the functional benefits of a 12-week tailored, structured, and progressive multidomain physical RI program in older adults, following a HF admission, could be estimated by pre-intervention levels of proteins. MethodsHospitalized adults age [≥] 60 years with HF were randomized to 12-week multidimensional (aerobic, strength, balance, flexibility) physical therapist-led RI versus attention control (AC). Outcomes included change in the short physical performance battery (SPPB) and 6-minute walk distance (6MWD). At pre-intervention and 12-weeks, 92 circulating proteins associated with CVD and inflammation were measured. Using linear regression, LASSO and ensemble tree-based approaches for flexible feature selection, proteins were selected for stratification of treatment response. Decision trees were developed to incorporate a multi-protein approach to optimize stratification of treatment response (effect measure modifiers [EMM]) for both functional outcomes. Differences in 12-week temporal changes based on RI vs AC and mediation analysis were used to assess possible causal roles for these proteins. ResultsBaseline proteomic data was available in 243/349 participants (69.6%). For SPPB, 2 proteins (GP6, ST2), and for 6MWD, 3 proteins (ALCAM, Gal-4, PECAM-1) were selected as EMMs. For SPPB, a decision tree with ST2 followed by GP6 level identified a differential 12-week improvement in score for RI vs AC from -0.8 (95%CI -1.9, 0.2) to 2.1 (95%CI 1.5, 2.7). For 6MWD, the optimized decision tree was based only on ALCAM with a differential 12-week improvement in score for RI vs AC from 10m (95%CI -17m, 36m) to 85m (95%CI 56m, 113m). However, none of the EMM proteins were mediators of change for the functional outcomes. ConclusionsCirculating proteins related to CVD and inflammation (GP6, ST2, ALCAM) stratify a successful functional response to the REHAB-HF rehabilitation program in older adults following HF hospitalization. Further investigation is needed to identify proteins associated with a causal role for rehabilitation benefits.
Matching journals
The top 11 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Metabolic Modulation Predicts Heart Failure Tests Performance 93%
- Predicting 30-Day and 1-Year Mortality in Heart Failure with Preserved Ejection Fraction (HFpEF) 93%
- The impact of outdoor walking interventions on frailty among older adults with mobility limitations: findings from the Getting Older Adults Outdoors (GO-OUT) study 92%
Similar papers in this journal
- Walking pace optimizes conventional cardiovascular disease risk prediction models among vulnerable subpopulations: a prospective cohort study 93%
- Olfactory impairment and the risk of major adverse cardiovascular outcomes in older adults 93%
- Carotid Intima-media thickness, carotid distensibility and incident heart failure in older men: The British Regional Heart Study 92%
Similar papers in this journal
- Artificial Intelligence Methods to Detect Heart Failure with Preserved Ejection Fraction (AIM-HFpEF) within Electronic Health Records: An equitable disease prediction model 90%
- Hypoferremia reduces long-term risk of major adverse cardiovascular events after STEMI by averting the myocardial reactive iron storm 90%
- Development and validation of a risk prediction algorithm for high-risk populations combining genetic and conventional risk factors of cardiovascular disease 89%
Similar papers in this journal
- Dynamic Predictive Accuracy of Electrocardiographic Biomarkers of Sudden Cardiac Death within a Survival Framework: The Atherosclerosis Risk in Communities (ARIC) study 92%
- Proteomic signatures and machine learning based-prediction models for cardiovascular risk in survivors of myocardial infarction 91%
- Postoperative glycemic variability as a predictor for one-year mortality following coronary artery bypass grafting: A retrospective cohort study 90%
Similar papers in this journal
- Development and Validation of a parsimonious AI-Based Risk Score for Mortality in Heart Failure: A UK cohort study 92%
- Advanced ECG heart age estimation applicable to both sinus and non-sinus rhythm associates with cardiovascular risk, cardiovascular morbidity, and survival 90%
- Using ECG Machine Learning for Detection of Cardiovascular Disease in African American Men and Women: the Jackson Heart Study 90%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.