Atrial fibrillation after cardiac surgery: identifying candidate predictors through a Delphi process

Area of Science:

• Cardiology
• Cardiac Surgery
• Medical Informatics

Background:

• Atrial fibrillation after cardiac surgery (AFACS) is a frequent complication with significant negative impacts on patient outcomes.
• Accurate prediction of AFACS is crucial for risk stratification and implementing targeted preventive strategies.
• Existing prediction tools may benefit from a broader range of identified predictors.

Purpose of the Study:

• To systematically identify potential predictors of atrial fibrillation after cardiac surgery (AFACS) through expert consensus.
• To supplement existing data from systematic reviews and cohort studies for developing robust AFACS prediction models.
• To inform the PARADISE project (NCT05255224) by establishing a comprehensive list of candidate AFACS predictors.

Main Methods:

• A modified Delphi consensus process involving an international multidisciplinary expert panel.
• A two-stage approach: initial generation of a comprehensive list of variables, followed by refinement through voting.
• Variables were retained if selected by at least 40% of the panel members.

Main Results:

• A panel of 15 experts (including physicians, surgeons, nurses, pharmacists, and patient representatives) participated.
• The Delphi process generated a final list of 72 candidate AFACS predictors.
• These predictors encompass demographic, comorbidity, vital sign, intraoperative, and postoperative factors.

Conclusions:

• The Delphi consensus method effectively identified a wide array of candidate AFACS predictors, potentially extending beyond current literature.
• The identified predictors provide a foundation for enhancing the validity and scope of AFACS prediction tools.
• These findings are integral to the ongoing development of AFACS prediction models within the PARADISE project.