Atrial Fibrillation, a Geriatric Giant

The studies in the thesis apply opportunistic case-finding strategies for atrial fibrillation, and addresses physicians’ worries about anticoagulation and major bleeding. Within the range of cross-sectional data from the Fall and Syncope Registry, through the single centre prospective case finding study, and then the multi centre case finding GERAF study, we consistently show that the prevalence and incidence of atrial fibrillation is very high among geriatric patients. With a baseline prevalence between 20 and 23%, and detection rate between 2.7 and 5.5%, it is roughly 1.5 times as likely that a patient at the geriatric outpatient clinic has atrial fibrillation, than that they have a definite diagnosis of dementia. Atrial fibrillation showed to be intimately intertwined with frailty, polypharmacy, multi morbidity, falls, gait disturbance, heart failure and other morbidities. Old age atrial fibrillation should be considered a geriatric feature, and all geriatricians need to consider its treatment, including stroke prevention, as usual day-to-day care. In the single lead ECG screening study, a striking incidence of 5.5% new atrial fibrillation was detected, with only 1 baseline ECG and an average of 3 single lead ECGs per patient, making it a very attractive method for in-office case-finding. In the Dutch-GERAF study, patients could screen at home, using their own smartphone. However, a confirmatory ECG remained necessary for a definite diagnosis of atrial fibrillation. Geriatrics is both about protecting frail patients to excessive treatment and diagnostics, as addressing over-cautiousness, for example the safety concerns for anticoagulation. In 2010 up to 60% of frail patients with atrial fibrillation received anticoagulation, with a perceived high risk of bleeding as the main argument to withhold anticoagulation. This shifts towards much higher rates of treatment, as reported in the MyDiagnostick study, 71%, and in the GERAF study 84%, and initiation of novel oral anticoagulation (NOAC) in 90% of the newly detected cases. The analyses on major bleeding events, and mortality associated with OAC, focus on the assumption that frail patients are more prone to OAC related major bleeding, especially if they also fall. The results presented in Chapter III however support this assumption to no extent at all, as they show that within the Fall and Syncope cohort the rate of major bleeding seems either equal (in case of the intracranial haemorrhage), or lower (in case of the other major bleedings), and that less than 10% of major bleedings are the consequence of a fall of accident, of which none were intracranial haemorrhages. Also, the use of VKA or antiplatelet agents were not independently associated with mortality. The mortality rate of the cohort is, as expected for frail older patients, very high, close to 3.8% per year, and 30% overall in the 8 year observation period. Only a small percentage of 6% concerns fatal bleedings. It is very likely that competing risks explain the lower rate of major bleeding in the Fall and Syncope Registry, as compared to the RCTs. The answer to the question if it is risky to treat a frail older patient is: ‘Yes, the risk of the ultimate adverse event, death, is quite high’. But at the same time the answer is also, ‘No, there seems not to be an exaggerated risk for anticoagulation related major bleeding’. The third answer summarises both into the insight: ‘Falls can help to identify frail patients, who are at an increased risk of adverse events, such as major bleeding and mortality. Bleeding as a direct consequence of a fall however was rarely observed, and therefore falls should be viewed as a biomarker for frailty and increased risk for adverse events, but not to be confused with causality’.