Patient selection for cardiac surgery: Time to consider subgroups within risk categories?

Area of Science:

• Cardiology
• Medical Statistics
• Health Services Research

Background:

• Medical guidelines often assume risk categories are homogeneous, overlooking patient variability.
• Individuals with similar predicted risks may require different treatment approaches.
• Tailoring treatment allocation necessitates identifying distinct patient subgroups.

Purpose of the Study:

• To evaluate a strategy for identifying homogeneous subgroups within risk categories.
• To empirically illustrate this strategy for high-risk cardiac surgery patients.
• To enable more precise treatment allocation based on identified subgroups.

Main Methods:

• Cluster analysis applied to a dataset of 6517 cardiac surgery patients.
• Analysis performed separately for younger (<75 years) and older (≥75 years) patient groups.
• High postoperative mortality risk defined by EuroSCORE ≥ 15%.

Main Results:

• Two distinct, robust subgroups identified in younger patients, differing in preoperative state and recent cardiac events.
• Two distinct, robust subgroups identified in older patients, differing in preoperative state and comorbidities like COPD and prior surgery.
• Validity measures confirmed the quality and robustness of identified subgroups.

Conclusions:

• A feasible method for identifying homogeneous subgroups within risk categories was demonstrated.
• Subgroup-specific treatment strategies can replace average-risk strategies, advancing individualized care.
• Focusing on specific subgroups can improve allocation discussions for cardiac surgery interventions.