The KARDIA BAND is a novel technology that enables patients to record a rhythm strip using a Connected Watch. The band is paired with an app providing automated detection of ATRIAL FIBRILLATION.
The purpose was to examine whether the KARDIA BAND could accurately differentiate SINUS RHYTHM from ATRIAL FIBRILLATION compared with physician-interpreted 12-lead ELECTROCARDIOGRAMs and KARDIA BAND recordings.
Consecutive patients with ATRIAL FIBRILLATION presenting for CARDIOVERSION were enrolled. Patients underwent pre-CARDIOVERSION ELECTROCARDIOGRAM along with a KARDIA BAND recording. If CARDIOVERSION was performed, a post-CARDIOVERSION ELECTROCARDIOGRAM was obtained along with a KARDIA BAND recording. The KARDIA BAND interpretations were compared to physician-reviewed ELECTROCARDIOGRAMs. The KARDIA BAND recordings were reviewed by blinded electrophysiologists and compared to ELECTROCARDIOGRAM interpretations. Sensitivity, specificity, and K coefficient were measured.
A total of 100 patients were enrolled (age 68 ± 11 years). Eight patients did not undergo CARDIOVERSION as they were found to be in SINUS RHYTHM. There were 169 simultaneous ELECTROCARDIOGRAM and KARDIA BAND recordings. Fifty-seven were noninterpretable by the KARDIA BAND. Compared with ELECTROCARDIOGRAM, the KARDIA BAND interpreted ATRIAL FIBRILLATION with 93% sensitivity, 84% specificity, and a K coefficient of 0.77. Physician interpretation of KARDIA BAND recordings demonstrated 99% sensitivity, 83% specificity, and a K coefficient of 0.83. Of the 57 noninterpretable KARDIA BAND recordings, interpreting electrophysiologists diagnosed ATRIAL FIBRILLATION with 100% sensitivity, 80% specificity, and a K coefficient of 0.74. Among 113 cases where KARDIA BAND and physician readings of the same recording were interpretable, agreement was excellent (K coefficient = 0.88).
As a conclusion, the KARDIA BAND algorithm for ATRIAL FIBRILLATION detection supported by physician review can accurately differentiate ATRIAL FIBRILLATION from SINUS RHYTHM. This technology can help screen patients prior to elective CARDIOVERSION and avoid unnecessary procedures.
Atrial fibrillation (ATRIAL FIBRILLATION) is the most commonly encountered arrhythmia in clinical practice and population-based studies forecast over 6 million individuals living with this diagnosis by 2050 1, 2. It is a chronic condition whose prevalence increases with age, and represents a growing economic burden for our health care system 3, 4. Although the journey of ATRIAL FIBRILLATION begins with an initial diagnosis, its management is long term, nuanced, and often involves hospital-based interventions along the way, including electrical cardioversion (CARDIOVERSION).
Recently, commercially available handheld cardiac rhythm recorders have been developed that can record a rhythm strip using smartphone technology (5). In November 2017, the Kardia Band (KARDIA BAND) (AliveCor, Mountain View, California) was introduced as the first U.S. Food and Drug Administration (FDA)–cleared Apple Watch accessory (Apple, Cupertino, California) that allows a patient to record a rhythm strip equivalent to lead I for 30 s. The KARDIA BAND is coupled with an application that provides an instantaneous and automatic rhythm adjudication algorithm for the diagnosis of ATRIAL FIBRILLATION. The application can inform the patient when ATRIAL FIBRILLATION is detected and transmit these results to the patient’s caring physician instantaneously.
The primary objective of our study was to examine whether the KARDIA BAND and ATRIAL FIBRILLATION detection algorithm could accurately and reliably differentiate sinus rhythm (SINUS RHYTHM) from ATRIAL FIBRILLATION when compared with physician-interpreted 12-lead ELECTROCARDIOGRAMs and KARDIA BAND recordings in patients with known ATRIAL FIBRILLATION presenting to a high-volume hospital-based electrophysiology practice for scheduled electrical CARDIOVERSION.
This was a prospective, nonrandomized, and adjudicator-blinded study completed at a tertiary care hospital-based electrical CARDIOVERSION laboratory designed to evaluate the accuracy of the KARDIA BAND automated algorithm for the detection of ATRIAL FIBRILLATION. AliveCor provided the KARDIA BAND connected to an Apple Watch which was paired via Bluetooth to a smartphone device (Apple) for utilization in the study (Figure 1). The Cleveland Clinic’s Institutional Review Board approved the study.
Consecutive patients with a diagnosis of ATRIAL FIBRILLATION who presented for scheduled elective CARDIOVERSION with or without a planned transesophageal echocardiogram were screened for enrollment. Inclusion criteria included all adult patients age 18 to 90 years who were able to provide informed consent and willing to wear the KARDIA BAND before and atrial fibrillationter CARDIOVERSION. We excluded all patients with an implanted pacemaker or defibrillator.
Once enrolled, patients underwent a pre-CARDIOVERSION ELECTROCARDIOGRAM followed immediately by KARDIA BAND recording. These paired recordings were considered simultaneous. If the CARDIOVERSION was performed, a post-CARDIOVERSION ELECTROCARDIOGRAM was then obtained along with another KARDIA BAND recording. The KARDIA BAND tracing was automatically analyzed using the KARDIA BAND algorithm. This algorithm measures rhythm irregularity and P-wave absence in real time to classify the rhythm strip as “possible ATRIAL FIBRILLATION.” If the criteria for ATRIAL FIBRILLATION is not met, the KARDIA BAND algorithm classifies regular rhythms with P waves as “normal” if the rate is between 50 and 100 beats/min or “unclassified” for those rhythms with rates <50 or >100 beats/min or if the recording is noisy or shorter than 30 s. The KARDIA BAND rhythm strips were automatically transferred to the secure AliveCor server, downloaded, and printed for review.
All automated KARDIA BAND rhythm strips and ELECTROCARDIOGRAMs were anonymized and distributed to 2 blinded electrophysiologists (BB and DC) who independently interpreted each tracing and assigned a diagnosis of SINUS RHYTHM, ATRIAL FIBRILLATION or atrial flutter, or unclassified. If the 2 electrophysiologists disagreed on the diagnosis, a third electrophysiologist (AH) reviewed the tracing and assigned a final diagnosis. To assess the accuracy of the KARDIA BAND algorithm at appropriately identifying ATRIAL FIBRILLATION, the automated KARDIA BAND interpretations were compared with both the physician-interpreted KARDIA BAND rhythm strips and the physician-reviewed simultaneous ELECTROCARDIOGRAMs.
Sensitivity and specificity were calculated for KARDIA BAND automated interpretation compared with physician-interpreted 12-lead ELECTROCARDIOGRAM, for physician interpreted KARDIA BAND rhythm strip compared with physician-interpreted 12-lead ELECTROCARDIOGRAM, and for KARDIA BAND automated interpretation compared with physician-interpreted KARDIA BAND recordings. Kappa (?) coefficients for interobserver agreement were assessed. ? coefficients >0.80 were considered to represent excellent agreement. ATRIAL FIBRILLATION and atrial flutter were considered a single disease state for all interpretations.
A total of 100 patients were enrolled in the study from March 2017 through June 2017. Demographics and clinical characteristics are summarized in Table 1. CARDIOVERSION was performed in 85% of study participants. Of the 15 patients who did not undergo CARDIOVERSION, 8 were cancelled due to presentation in SINUS RHYTHM. There were 169 simultaneous 12-lead ELECTROCARDIOGRAM and KARDIA BAND recordings obtained from study participants, and 57 KARDIA BAND recordings were determined as unclassified by the KARDIA BAND algorithm. Of the 57 unclassified KARDIA BAND tracings, 16 (28%) were due to baseline artifact and low amplitude of the recording, 12 (21%) were due to a recording of <30 s in duration, 6 (10%) were due to a heart rate of <50 beats/min, 5 (9%) were due to a heart rate of >100 beats/min, and the remaining 18 (32%) were unclassified due to an unclear reason. Electrophysiologist-interpreted 12-lead ELECTROCARDIOGRAMs were all interpretable.
To test the ability of the KARDIA BAND algorithm to detect ATRIAL FIBRILLATION, automated KARDIA BAND rhythm interpretations and electrophysiologist-interpreted 12-lead ELECTROCARDIOGRAMs were compared. Among the recordings where the KARDIA BAND provided a diagnosis, it correctly diagnosed ATRIAL FIBRILLATION with 93% sensitivity, 84% specificity, and a K coefficient of 0.77 (95% confidence interval: 0.65 to 0.89) when compared with the electrophysiologist-interpreted 12-lead ELECTROCARDIOGRAM (Table 2). Because our analysis used multiple observations from the same individual, we evaluated for possible intraindividual correlations by comparing only pre-CARDIOVERSION KARDIA BAND recordings to electrophysiologist-interpreted 12-lead ELECTROCARDIOGRAMs and found the performance of the KARDIA BAND algorithm to be unchanged.
To determine whether the automated KARDIA BAND recordings labeled as “unclassified” by the algorithm were still clinically useful, these tracings were interpreted by our blinded electrophysiologists and compared with the electrophysiologist-interpreted 12-lead ELECTROCARDIOGRAMs. Of the 57 automated unclassified KARDIA BAND recordings, the interpreting electrophysiologists were able to correctly diagnose ATRIAL FIBRILLATION with 100% sensitivity, 80% specificity, and a K coefficient of 0.74.
To assess the fidelity and overall quality of the KARDIA BAND tracings produced by the smartwatch, electrophysiologist-interpreted KARDIA BAND recordings were compared to corresponding 12-lead ELECTROCARDIOGRAM tracings. A total of 22 recordings were determined to be noninterpretable by the reading electrophysiologist, and these were predominately due to baseline artifact. Of the remaining 147 simultaneous recordings, the electrophysiologist interpreted 12-lead ELECTROCARDIOGRAMs and electrophysiologist interpreted KARDIA BAND recordings, physician interpretation of the KARDIA BAND tracings demonstrated 99% sensitivity, 83% specificity and a K coefficient of 0.83
Additionally, to measure the quality of the KARDIA BAND recordings, we compared the KARDIA BAND automated algorithm interpretation to physician interpretation of the same recordings. Of the cases where both methods were interpretable, the KARDIA BAND automated algorithm was 93% sensitive and 97% specific in detecting ATRIAL FIBRILLATION with a K coefficient of 0.88
The era of mobile health care technology has proliferated over the past decade. Consumers from the general public now have direct access to devices and applications that offer real-time measurements of cardiovascular physiology, and some technologies extrapolate this data to provide diagnostic information (6). It is estimated that by 2019, annual sales of such devices will reach 50 billion dollars worldwide (7). However, the ability of some devices to accurately measure biometric endpoints has been questioned, and some mobile health technologies are available without verification through rigorous clinical studies (8).
Alongside the growth of mobile health care technology has been the desire of many physicians and patients to accurately monitor disease-related metrics of chronic conditions in the ambulatory setting. ATRIAL FIBRILLATION is a good example of a relapsing condition that requires frequent monitoring of clinical endpoints to assess the efficacy of treatment choices and plan future interventions. The KARDIA BAND is the first smartwatch accessory cleared by the FDA and available to the general public without a prescription that claims to instantaneously detect ATRIAL FIBRILLATION and transmit this information to a patient’s treating physician.
In this study, we aimed to assess whether the KARDIA BAND and ATRIAL FIBRILLATION detection algorithm could accurately and reliably differentiate SINUS RHYTHM from ATRIAL FIBRILLATION in patients with known ATRIAL FIBRILLATION presenting for scheduled electrical CARDIOVERSION (Central Illustration). We compared automated KARDIA BAND interpretations to simultaneously recorded ELECTROCARDIOGRAMs read by blinded electrophysiologists and found very good agreement between them. When able to provide an interpretation, the automated KARDIA BAND readings correctly identified ATRIAL FIBRILLATION with 93% sensitivity and 84% specificity (Figure 2). Of the 169 total KARDIA BAND recordings, 57 (33.7%) were interpreted as unclassified by the automated KARDIA BAND algorithm. Reasons that these recordings were deemed noninterpretable included short recordings <30 s, low-amplitude P waves, and baseline artifact. For those recordings where the automatic KARDIA BAND tracing was noninterpretable, direct physician interpretation could be used to correctly identify ATRIAL FIBRILLATION with 100% sensitivity and 80% specificity (Figure 3). In general, the KARDIA BAND recordings when interpreted by the physician had excellent agreement with simultaneous 12-lead ELECTROCARDIOGRAM interpretation with 99% sensitivity and 83% specificity.
Prior to the development of the KARDIA BAND smartwatch algorithm, several algorithms used by implantable loop recorders (ILRs) were validated for the detection of ATRIAL FIBRILLATION. Currently available ILRs detect ATRIAL FIBRILLATION by sensing R waves and applying a variety of regularity algorithms to detect ATRIAL FIBRILLATION. The Confirm DM2101 (Abbott, Chicago, Illinois) detects RR interval regularity and measures suddenness of an irregular rhythm’s onset and offset to diagnose ATRIAL FIBRILLATION using 2 probabilistic scoring models. The BioMonitor (Biotronik, Berlin, Germany) also measures R-wave variability and allows the clinician to adjust the number of cycle lengths used and the confirmation time needed to detect ATRIAL FIBRILLATION. The most studied of the ILRs is the Reveal LINQ (Medtronic, Minneapolis, Minnesota) system whose algorithm for ATRIAL FIBRILLATION detection uses both R-wave irregularity and a programmable P-wave evidence discrimination tool that can be modified based on the individual needs of a given patient 9, 10, 11. The Reveal LINQ system was evaluated in the XPECT (Reveal XT Performance Trial). In this study, the sensitivity and specificity for identifying patients with any ATRIAL FIBRILLATION was 96.1% and 85.4%, respectively (12). In our study, the accuracy of the KARDIA BAND algorithm for the detection of ATRIAL FIBRILLATION was comparable to these results.
Wearable devices like the KARDIA BAND require a satrial fibrillatione and durable platform upon which recordings can be reviewed and stored. A secure cloud-based platform has been developed to view and download KARDIA BAND recordings. The applicability of this platform to the outpatient management of patients with ATRIAL FIBRILLATION needs to be evaluated and studied in future trials. Our study also demonstrated that a subset of patients (8%) who presented for CARDIOVERSION was found to be in SINUS RHYTHM. For each of these patients, the automated KARDIA BAND algorithm did not erroneously identify ATRIAL FIBRILLATION, and the physician interpretation of the KARDIA BAND recording correctly confirmed SINUS RHYTHM in each case. Although this study was not powered to assess the financial consequences of cancelled CARDIOVERSIONs, it is reasonable to conclude that a measurable number of resources were forfeited by both the patient and the health care system in anticipation of a procedure that was ultimately deemed unnecessary once SINUS RHYTHM was confirmed. As data from the KARDIA BAND can be reviewed remotely, the resources used in preparation of these patients’ cancelled CARDIOVERSIONs could have been saved. The KARDIA BAND system has been previously shown to be cost-effective for ATRIAL FIBRILLATION screening. Our study suggests the potential use of these products to provide more effective health care delivery (13).
This was a single-center study at a tertiary referral center with a small sample size. The population represented in this study had a known history of ATRIAL FIBRILLATION and a sufficient burden of ATRIAL FIBRILLATION to prompt electrical CARDIOVERSION. The performance of the KARDIA BAND smartwatch algorithm may be more variable in a population with a lower ATRIAL FIBRILLATION burden. We did not evaluate socioeconomic status in our study, and only 17% of our enrolled patients were female. Additionally, none of the patients who participated in our study had previously used the KARDIA BAND. These facts may limit the generalizability of our findings in the general public, and future studies should consider measuring these variables. Patients with cardiac implantable electronic devices were excluded from this study, and further evaluation of the KARDIA BAND algorithm is needed in this patient population. Participants were instructed on how to use the KARDIA BAND wristband while seated in a hospital bed immediately prior to obtaining each recording. Their ability to record each tracing was directly observed. As a result, the performance of the KARDIA BAND algorithm and the clarity of the recorded tracings may be less accurate in an outpatient or ambulatory setting. For the same reason, some of the unclassified recordings could have been avoided with more patient practice on the proper use of the KARDIA BAND device. Additionally, the KARDIA BAND prototype used in our study did not display a real-time ELECTROCARDIOGRAM tracing on the watch screen at the time of recording. Since FDA clearance, the KARDIA BAND app is now permitted to display this information. We anticipate the real-time display of the ELECTROCARDIOGRAM recording will improve the quality of the recordings obtained by users of the device.
The KARDIA BAND smartwatch automated algorithm for ATRIAL FIBRILLATION detection, supported by physician review of these recordings, can reliably differentiate ATRIAL FIBRILLATION from SINUS RHYTHM. Avoiding scheduling unnecessary electrical CARDIOVERSIONs is 1 example of a clinical application of the KARDIA BAND system. Many other potential applications warrant further investigation and might transform our longitudinal care of ATRIAL FIBRILLATION patients.
Source: Science Direct