Description of the health problem
New-onset atrial fibrillation (NOAF) is defined as atrial fibrillation (AF) that occurs in a patient with no known history of chronic or paroxysmal AF.1 It is a common arrhythmia in critically ill patients.2 It occurs in 5–15% of all patients admitted to a general intensive care unit (ICU),3,4 rising to 23% of patients with septic shock.5
Organised atrial activity is important for ventricular filling and cardiac output.6 NOAF is temporally associated with a reduction in cardiac output in non-ICU patients.7 The haemodynamic impact of NOAF in critically ill patients is poorly understood, but limited data suggest that NOAF may precede haemodynamic instability8 and may be associated with increased rates of thromboembolism.9 NOAF during critical illness is associated with an increased risk of death in an ICU and in hospital.10,11 There is also a significant organisational impact of NOAF because it is associated with an increased length of ICU and hospital stay, and higher health-care costs.12
New-onset atrial fibrillation during critical illness may carry a long-term burden. Patients who develop NOAF during sepsis and survive to hospital discharge have an increased risk of heart failure and stroke, and poorer 1-year and 5-year survival.13,14 The long-term outcomes for patients who develop NOAF in an ICU remains unclear.
It is not known whether NOAF in patients in an ICU is causally related to worse outcomes or whether NOAF may be solely a marker of disease severity. However, there is clear mechanistic plausibility behind a causal association. This demonstrates the need for optimal prevention, management and follow-up. Although a recent scoping review has broadly described studies of NOAF treatment in patients in an emergency department or an ICU, or after major surgery,10 an in-depth review of NOAF in patients in an ICU focusing on treatment efficacy is required to put current treatment practices into context and to inform future comparative studies.
Clear guidelines exist for the management of AF in patients in the community.15 However, there is a paucity of evidence for its management in the critical care setting, for which the balance of risks and benefits associated with different treatment options is unclear. Understandably, there is significant variation within and between units in the management of this common problem.16 Many previous studies informing NOAF treatment in ICUs are small or inadequately adjusted for confounding factors. Well-conducted, multicentre, observational studies are required to highlight candidate interventions for clinical trials.
Overall aims and objectives of the study
Scoping review
- To evaluate the evidence for the clinical effectiveness and safety of pharmacological and non-pharmacological NOAF treatments.
- To provide guidance for the database analysis on:
- NOAF definitions used for patients in an ICU
- patient subgroups who develop NOAF in an ICU
- inclusion/exclusion of specific treatments and potential confounders
- determining barriers to future research.
Database analysis: RISK-II
- To determine how common NOAF is in critical care.
- To determine the typical characteristics of patients with NOAF in critical care and how they compare with other patients in critical care.
- To increase the understanding of the outcomes of patients with NOAF in critical care and how they compare with other patients in critical care.
- To investigate how much of the difference in outcomes is explained by differences in patient characteristics and comorbidities.
Database analysis: MIMIC-III and PICRAM
- To compare the use and clinical effectiveness of pharmacological and non-pharmacological NOAF treatments.
- To determine the incidence of short- and long-term NOAF complications.
Patient and public involvement
Patient and public involvement was vital throughout the study. Valerie Keston-Hole and Rob Lawrence helped to develop the application, which was also reviewed by the Oxford Critical Care Patient Forum. The group strongly supported the use of existing databases for the purposes of undertaking the work. Valerie Keston-Hole sits on the group, which assesses applications for the use of the Post Intensive Care Risk-adjusted Alerting and Monitoring (PICRAM) data used in this work. Ian and Cathy Taylor provided us with a clear patient perspective when working with the expert panel and also helped us to choose research recommendations. Meetings went well and easy access to the chief investigator meant that things that were unclear could be explained by e-mail afterward and further thoughts considered. In discussion with Ian and Cathy Taylor, we identified that an area that we would improve in the future was how to present large numbers of initial data in a more comprehensible manner to our patient and public involvement (PPI) colleagues. A suggestion for the future would be to provide supplementary information that avoided technical terminology to help the understanding of the data by our PPI colleagues prior to the meetings. This would allow more spontaneous comments and discussion during the meeting. Our PPI work is not complete. We discussed our findings at the ICU patient forum. This helped us to understand how to clearly communicate our findings.