What is this study about?
Atrial fibrillation is a heart problem causing a fast, irregular heartbeat. This reduces the heart’s ability to pump blood around the body. It also causes blood clots to form inside the heart. These can spread through the blood vessels to other parts of the body. These clots can cause strokes if they spread to the brain.
Atrial fibrillation is a common problem in patients outside intensive care units (ICUs) and good, evidence-based, guidelines exist to help doctors treat people who develop this condition. Around 10% of people treated on an ICU develop atrial fibrillation as a complication of their severe underlying illness. This additional problem makes them more unstable, so they stay longer in the ICU and have worse outcomes. Atrial fibrillation therefore needs prompt and effective treatment to prevent further harm.
Treatments for atrial fibrillation that work in people outside the ICU may not work in people treated on an ICU who are already very ill before their heart changes rhythm. This means that guidelines for treating atrial fibrillation outside ICU are not helpful for patients treated on an ICU. There is uncertainty about what is the best treatment and practices differ between countries, and between different ICUs in the same country.
People who have atrial fibrillation outside ICU often have their blood thinned to reduce their risk of stroke. However, this treatment can cause bleeding. Risk scoring systems are used to help doctors balance the risk of bleeding against the risk of stroke, but these scoring systems may not work for patients admitted to ICUs. This is because the risk of bleeding is higher, and the risk of stroke is not well understood.
Critical care Atrial Fibrillation Evaluation (CAFE) study will bring together the best evidence on which to base improved guidelines for treatment of patients who develop atrial fibrillation on an ICU.
What does the study involve?
We will start with a scoping review of all published research and expert opinion to establish the effectiveness of current treatment strategies for new onset atrial fibrillation in adults on an ICU. It will also suggest the best areas for future research.
To investigate the benefits and harms of existing treatments for atrial fibrillation, we will then use databases of medical records of patients treated on ICUs:
- PICRAM is a large detailed research database of patients admitted to three ICUs in the UK. It includes 18,000 patients.
- MIMIC-III holds similar data on 40,000 patients from the ICUs of two US hospitals. Many treatments are different in the two countries, including treatment for atrial fibrillation.
- The NIHR HIC critical care database holds some of the clinical details on 22,000 patients from five UK ICUs. We will check some of our findings with this dataset.
To understand whether we could improve outcomes for patients who develop atrial fibrillation on ICU by thinning their blood, we need to know how frequently strokes occur, both in hospital and after they go home. We will use the UK national RISK-II database of over half a million patients in 200 ICUs to estimate the incidence of stroke both in hospital and after discharge.
This study is funded by the National Institute for Health Research (NIHR) - Health Technology Assessment (HTA) (17/71/04)