Patients with ischemic cardiomyopathy had better mean quality of life (QoL) scores after regular coronary artery bypass surgery (CABG) compared to guideline-directed medical therapy (GDMT), according to the Surgical Treatment for Ischemic Heart Failure (STICH) trial. Patients wanting information about a high-risk/high-benefit treatment option need more than the average disparities in QoL ratings provide. To classify QoL data into clinically relevant result patterns (phenotype trajectories), researchers evaluated Kansas City Cardiomyopathy Questionnaire (KCCQ) Overall Summary scores in CABG and GDMT patients over a 36-month period using a mix of statistical approaches. Then they identified the key baseline predictors for each phenotype. To characterize the most common patterns in STICH QoL data, mixture models were used to construct phenotypes of QoL outcomes. The likelihood of each outcome pattern occurring for a given patient under each treatment was estimated using logistic regression models. There were 592 CABG procedures and 607 GDMT patients in STICH. By analyzing data from both therapy groups, researchers were able to determine 3 distinct phenotypic trajectory patterns. A total of 498  (84.1%) CABG patients and 449 (73.9%) GDMT patients experienced 1 of 2 improvement patterns (excellent QoL trajectories) out of the 3 possible trajectories. About  277 (23.5%) patients were more likely to have a good outcome with CABG, while 45 (3.8%) patients were more likely to have a good outcome with GDMT, using the definition of a consequential CABG-GDMT treatment difference as more than 10% higher absolute predicted probability of belonging to good QoL trajectories. The success rates for CABG and GDMT were within 5% of each other for 644 patients (54.7%). Individual baseline characteristics allowed researchers to predict 1 of the 3 main phenotypic trajectories in the pattern of QoL outcomes following CABG compared with GDMT in STICH. The ability to make informed decisions based on patient-specific forecasts of QoL outcomes under various treatment scenarios is a powerful tool in the field of personalized medicine.

Source: sciencedirect.com/science/article/abs/pii/S0002870322001600

Author