Smartphone-based device, PWV detect atrial fibrillation in primary care

The Kardia Mobile Cardiac Monitor (KMCM), a handheld, smartphone-coupled, two-electrode device for recording cardiac rhythm, can be used to reliably screen for atrial fibrillation (AF) in the primary care setting, according to a recent study. Similarly, pulse rate variability (PRV) may be used as an accurate marker for AF.

“AF can be detected by PRV accurately and by KMCM, especially when used in combination and in the absence of non-AF arrhythmias. Performing two PRV measurements reduces misclassification,” the researchers said. “Through early AF detection and subsequent treatment, implementing such measurements should have a meaningful impact on the adverse health consequences of AF.”

The study included 421 primary care patients, of whom 133 were definitively diagnosed with AF using the standard 12-lead electrocardiogram (ECG) approach. In parallel, participants were also fitted with a blood pressure (BP) device for the measurement of PRV, and with the KMCM for classification as “normal,” “possible AF,” or “unclassified.”

The BP device found that all PRV parameters were markedly higher in participants with AF than in those without: standardized average real variability (sARV; 16.2±7.4 percent vs 2.2±2.9 percent), root mean square of successive differences (RMSSD; 139±104 vs 23±34 ms), coefficient of variation (CV; 12.6±5.7 percent vs 2.1±2.7 percent), and irregular pulse period (IPP; 50.0±35.0 percent vs 9.1±11.0 percent) among others. [Sci Rep 2021;11;17721]

Discriminative performance analysis showed that PRV could reliably predict AF. Among its parameters, sARV provided the best discriminative value, with an area under the curve (AUC) of 0.94 when using the average of the first and second measurements. Corresponding values for RMSSD and CV were 0.92 and 0.93. Of note, IPP yielded the lowest predictive value, with an AUC of 0.87.

PRV remained a reliable marker of AF even in various subgroup analyses. In particular, excluding patient with AF or focusing on participants without pacemakers even slightly improved PRV’s predictive capacity.

Similarly, KMCM proved to be useful for AF detection. The smartphone-based device found 134 patients (of a total of 421 readings) to have possible AF; 198 were classified as normal, 72 could not be classified, and 17 were blank. Compared with ECG and even while retaining unclassified and blank readings, KMCM yielded sensitivity of 83 percent and specificity of 68 percent. Its diagnostic accuracy was 73 percent.

When excluding unclassified and blank readings from the analysis, sensitivity, specificity, and diagnostic accuracy all improved to 97 percent, 89 percent, and 92 percent, respectively.

“Finally, using combinations of measurements—two PRV ones or both a PRV and a KMCM recording—increased precision compared with when a single measurement was employed,” the researchers said.

“Intervention studies would clarify whether performing these measurements in clinical practice [could] lead to the expected improvements in AF-related health outcomes,” they added.