Neurology News

Follow our Neurology News blog for the latest news on neurologic diseases and research.

Thursday, July 24, 2014

Can Paroxysmal Atrial Fibrillation in Stroke Patients Be Detected at Home?

by Kurt Samson


With proper training, acute ischemic stroke (AIS) patients and/or their caregivers can learn to distinguish paroxysmal atrial fibrillation from normal heart rhythm by taking a simple wrist pulse measurement. The study, which was conducted at a 14-bed stroke unit in Germany, was published in the July 23 online edition of Neurology.  

Measured peripheral pulse (MPP) is the only technique recommended by international guidelines to detect paroxysmal atrial fibrillation (pAF), but many false-alarms could be prevented if it is done at home, according to Bernd Kallmünzer, MD, and colleagues at the Universitätsklinikum Erlangen, in Erlangen, Germany.

They tested the accuracy of home-monitored MPP in a prospective study involving 256 patients trained to take their wrist pulse and detect pAF using a stopwatch. The study included all AIS patients treated at a single tertiary care facility between April 2012 and February 2013. Data was compared with simultaneous blinded electrocardiogram (ECG) for diagnostic accuracy and outcomes were reviewed.

Patients and caregivers were taught to learn how to distinguish between normal or arrhythmic pulse indicating possible pAF. If two subsequent measurements were accurate after training, it was considered a success. Patients with known pAF were included only if they were unaware of their current heart rhythm at the outset and during MPP.

Prior studies have shown that pulse taking by health care professionals had repeatedly high sensitivity for AF detection, ranging from 92 percent 100 percent.

In this study, the accuracy in taking MPP by health care professionals or patient’s relatives had a diagnostic sensitivity of 96.5 percent and 76.5 percent, respectively, they found, with 94.0 percent and 92.9 percent specificity in detecting AF. Among competent patients, self-measurements were reliably performed by 89.1 percent, with a diagnostic sensitivity of 54.1 percent and 96.2 percent specificity. Further, the rate of false positives was low, in only 2.7 percent of cases, with a positive predictive value of 76.9 percent and a negative predictive value of 90.0 percent.

The sensitivity of 54.1 percent in detecting atrial fibrillation was derived from a single measurement, but may increase by repeated testing, the researchers noted, however false positive results were limited to a very low rate of 2.6 percent.

“Survivors of ischemic stroke are at high risk for paroxysmal atrial fibrillation, although the episodes are often asymptomatic,” Dr. Kallmünzer told Neurology Today. “Although the episodes are often asymptomatic, the patients with AF should be offered antithrombotic treatment to prevent cardioembolism and recurrent stroke.”

The major findings, he said, were that MPP by a health care professional has excellent sensitivity and specificity for pAF-detection after ischemic stroke, but that most stroke patients and their relatives can also learn the technique with a low rate of false-positive results.

“Taking peripheral pulse is easy, free, and non-invasive, but highly effective as a first-step screening tool,” said Dr. Kallmünzer. “Many patients can be educated to distinguish between normal and absolute arrhythmic pulse sensation although they suffer from neurologic deficits from their stroke, and for patients with more severe handicaps, family members may perform the pulse measurements with a remarkable high sensitivity and specificity.

“Notably, there were no cases of atrial flutter in this study, which may be missed by MPP in case of rhythmic atrioventricular conduction depending on the resulting pulse rate,” according to the researchers. Moreover, a history of AF or oral anticoagulation, and the subtype of AF (paroxysmal vs. persistent/permanent) could bias the accuracy of MPP by patients and relatives, even though a subgroup analysis of patients with detected and undetected AF episodes did not show this to be the case.

Stay tuned for the full-text story and discussion in the August 21 issue of Neurology Today. See our archives for previous coverage of AF and stroke: