Patients with Atrial Fibrillation at High Risk for Stroke No More Likely to Receive Medication than Patients at Lower Risk

A new study found no correlation between stroke risk among atrial fibrillation patients and the likelihood that they’d receive prescriptions for risk-reducing medication during hospitalization.

Researchers from Northwestern University, who presented their findings at the Society of Hospital Medicine’s 2014 annual meeting, performed a retrospective review of 16,106 patients released over a 34-month period from a large tertiary medical center with a primary or secondary diagnosis of atrial fibrillation.

The study team judged the stroke risk for each patient by calculating his or her score on the CHADS2 scale (Congestive heart failure, Hypertension, Age, Diabetes mellitus, Stroke or TIA or thromboembolism). The researchers then matched scores to prescription records.

Current clinical guidelines recommend no anticoagulation or antiplatelet agents for a CHADS2 score of 0 and either an antiplatelet agent or full-dose anticoagulation for a score of 1. According to the guidelines, all patients with a score greater than 1 should receive full anticoagulation, but because stroke risk rises with CHADS2 scores, the need for medication also rises with scores.

The Northwestern study, however, found that discharged patients with a score of 6 (the highest CHADS2 score) were slightly less likely to go home with a prescription than those with a CHADS2 score of 0.

Overall, 39.3% of the 911 patients with a CHADS2 score of 0 went with no stroke prophylaxis, as did 40.3% of the 1,393 who scored 1, 38.8% of the 1,593 who scored 2, 35.9% of the 1,199 who scored 3, 36.8% of the 570 who scored 4, 36.0% of the 381 who scored 5 and 41.2% of the 114 who maxed out at 6.

“Objective stroke risk does not correlate with increased use of prophylaxis and a large percentage of patients at highest risk are given no prophylaxis of any kind,” wrote the authors of the study, Hiren Shah, MD, and N.P. Christensen.

“Current barriers to prophylaxis may be the lack of everyday use of objective risk tools, the expectation that treatment will be started on an outpatient basis or the lack of time needed for hospitalists to discuss the anticoagulation decision with patients.”

But for patients with high CHADS2 scores, the need for medication is immediate.

According to one large study published in JAMA, moving from 0-6 on the CHADS2 scale increases the annual risk of stroke for a patient with atrial fibrillation from 1.9% to 2.8% to 4.0% to 5.9% to 8.5% to 12.5% to 18.2%. Other studies have found similar numbers.

Speaking at the Society of Hospital Medicine meeting, Shah said that concern about bleeding risks might have dissuaded hospital staff from prescribing anticoagulants for some patients. However, he also said that, for patients with annual stroke risks of 12% or 18%, treatment was generally the right option, even if the risk from bleeding were considerable.

Shah also suggested that hospitals might often simply forget to calculate CHADS2 scores for patients with atrial fibrillation, but he identified a systematic solution for that problem in the study’s conclusion.

“Systems reminders should be generated to encourage the use of risk stratification tools and hospitalists should start prophylaxis during hospitalization or communicate stroke risk on handoff to outpatient physicians at time of discharge.”