In recent years, the management of atrial fibrillation (AF) has rapidly evolved, with many more options available to achieve either rate or rhythm control. Recently, other interventional approaches have been developed to terminate AF returning the patient's rhythm to normal sinus. These new approaches include catheter-based and surgical procedures designed specifically to treat AF.1
However, determining the successful return of sinus rhythm and the medical management after ablation requires the selection of the correct diagnostic method. In May 2007, the Heart Rhythm Society (HRS) in conjunction with the Society for Thoracic Surgeons, the European Heart Rhythm Association, and the European Cardiac Arrhythmia Society developed a consensus statement on catheter and surgical ablation of AF.2 The purpose of the consensus statement was to provide recommendations for ablation programs.
The task force identified that it is critical for surgical programs to adopt a consistent method of follow-up and monitoring. This recommendation was made because of lack of standardization in the reporting on the success of surgical and catheter ablation procedures, with the majority of reports based on the use of only intermittent electrocardiography (ECG) as a diagnostic method for recurrence of AF. Currently, the HRS is recommending that the minimal acceptable duration for assessing rhythm status is 24 hours Holter monitoring postprocedure. They have defined a successful outcome as having no monitored incidents of AF or other atrial arrhythmias (atrial flutter or atrial tachycardia) that last >30 seconds while off all antiarrhythmic medications. The reader is referred to the HRS guidelines for a complete discussion of the specific time periods recommended to conduct the long-term monitoring.2
Therefore, choosing the right monitor for the determination of the success, postsurgical ablation procedure that meets the HRS guidelines for continuous monitoring especially for asymptomatic patients is imperative. The use of monitors for periods >24 hours is more reliable in capturing events of atrial arrhythmias especially when they are short and the patient is asymptomatic.3 However, it is in compliance with the HRS guidelines to have 24 to 48 hours Holter monitoring being reviewed by board-certified cardiologists. When an institution is unable to support their own Holter monitor center where board-certified cardiologists can read and interpret patient results or would rather have longer monitoring periods like us, it is imperative to select a commercial system that can support your needs. Therefore, the purpose of this article is to assist practitioners in determining which monitor will best meet their needs by comparing several currently available heart rhythm monitors on selected criteria. The criteria chosen for this review is based on our current experience in the development of our monitoring program using a commercially available product to determine the return to sinus rhythm rate after surgical ablation in accordance with the HRS guidelines. The criteria selected to perform the review include (1) ease of use for patients (compliance), (2) ability of the monitor to perform internal analysis and storage capability, (3) external support, (4) AF burden report, and (5) reporting of industry standards and sensitivity.
Definitions of Monitors Reviewed
A Holter monitor test is defined as a portable continuous monitoring of the electrical activity of the heart usually >24 hours (can be used for longer periods) and uses three to eight electrodes placed on the anterior chest wall. The heart rhythm electrical activity is stored on a “flash” card and then downloaded to a computer analysis program in which a detailed report is generated giving summary statistics for the entire time of monitoring. The software can also alert the final reader to areas that may need further analysis. A patient is asked to keep a diary of anything unusual, such as palpitations or exercising, that occurs during the time they are wearing the monitor. The diary can be helpful to the practitioner in determining whether there is any correlation between symptoms and any arrhythmias detected by the monitor.4,5
Long-term monitoring and event monitor also provide portable monitoring of patients. The monitors are configured differently and may be based on single or multiple channels (usually two). They are designed to store rhythm data for an extended period of time, and these data may be related or unrelated to patients' symptoms. Those that are reliant on the patient wearing the monitor to push a bottom to record the heart rhythm at the time of their symptoms are more limited because many of the events may be silent. There are different modes of storage and wireless or telephone transition to a monitoring center provided by the different companies. The long-term/event monitor is usually ordered for infrequent reports of symptoms and can be worn for an extended period of time although the most frequent time period is 30 days.6 In this article, we are going to discuss the different monitors.
CRITERIA FOR EVALUATION OF ARRHYTHMIA MONITORS
The ability of patients to comply with wearing the monitor for >24 hours is especially important when choosing a monitor. One of the biggest concerns for patients when wearing the monitor for >1 day is the ability to bathe. The ease of lead placement by the patients themselves affords the patient to bathe at their leisure and to perform other activities including exercising at their discretion. In addition, the monitor should be able to provide a battery that lasts at least as long as the monitoring period. Many patients especially the elderly have found it to be a struggle for them to change batteries on regular basis. The size of the monitor is also important. Patients have reported that they will not wear the monitor if it is big, bulky, and not easily concealed beneath their clothes.
Internal Analysis and Storage Capability
Although previous research suggests that snap shots of the cardiac status provided by ECG may be insufficient in capturing the patient's true cardiac status, the HRS guidelines suggest at least 24 to 48 hours of monitoring. This requirement makes it essential that the monitor have the ability to perform an auto analysis of a patient's current rhythm and can trigger a response for a potentially life-threatening rhythm regardless of patient symptomology and activation. In addition, when monitoring for AF, it is important that the monitor must be able to analyze for R to R variability and p wave morphology allowing the monitor the capability of detecting asymptomatic atrial arrhythmias. It is also imperative that the monitor should be able to store rhythm information for at least the selected length of the monitoring period without having to erase any previous recordings. The summary of the report should be also evaluated to ensure the correctness of the report.
Depending on the size of one's practice, it may be very important to have an external support to operate a long-term monitoring program after ablation. The external support that will be necessary includes the ability of the monitoring company to send all equipment to the patients, file the insurance claims, and be available to patients for assistance wearing the monitors or contacting the patients if there are problems with the monitor. In addition, the ability for the monitoring company to provide a continuous analysis of transmitted rhythms, to identify life-threatening arrhythmias, to offer a practice to develop the ability of their own emergency notification criteria, and to call and direct patients when an emergency is noted as vital to a program.
Another very essential feature that external support programs can provide is a comprehensive written analysis with examples of the rhythms downloaded/transmitted from the monitor. Further, the ability to quantify AF burden is imperative to programs to determine their success rate. The AF burden reports should be able to quantify the total amount of time spent in AF and give by specific time periods for each incident of monitored AF (eg, number of incidents of AF <10 seconds, number of incidents <30 seconds, number of incidents >30 seconds, number of incidents <60 seconds). This type of reporting is essential for the small practices that do not have the resources to purchase their own monitors or the analytic software and the personnel needed to staff the monitoring center.
Currently, the industry standards for cardiac monitoring are captured in the ANSI/AAMI EC 57: 1998/2003 documents.6–9 These standards are a set of recommendations for testing cardiac rhythm systems and ST measurements and reporting the results of those tests, so that automated testing methods are reproducible. In addition, the industry standards for a device to monitor an ECG and to produce a visible or audible signal or an alarm when an arrhythmia is detected are found in the Association for the Advancement of Medical Instrumentation Electrography (AAMI EC) 38 document.9
In addition, the heart rate detector algorithms should also be tested against an ECG database.8 The three prominent databases are the (1) Massachusetts Institute of Technology-Beth Israel Hospital Arrhythmia Database (MIT DB), (2) American Heart Association Database for Evaluation of Ventricular Arrhythmia Detectors (AHA DB), and (3) Noise Stress Test database. The MIT and AHA databases use real ECGs for comparison testing and are used for determining the sensitivity and positive predictivity. The NST database has added artificial noise to the ECGs which is used to test how sensitive the monitors are for triggering on noise. The triggering on noise can usually be avoided with good skin preparation and high quality electrodes.6–9 However, reliance on the report received from the company is not advised, and manual review of the data by a cardiologist and or electrophysiologist should be completed to ensure the validity of the report.
SELECTED ARRHYTHMIA MONITORS
The following is a review of several of the monitors on the market using the above criteria. However, this review is not meant to be inclusive of all monitors. The review is based on the literature obtained from the companies and may not necessarily cover all the monitors features or what may be available in the future; the authors would refer the readers to the monitor companies for complete information. The purpose of the review is only to provide the practitioner with an overview of the monitors and criteria that may be considered when evaluating monitoring equipment for their practice (Table 1).
eCardio Diagnostics, The Woodlands, TX
eCardio provides both event and Holter monitors. Their event monitor (eTrigger AF920) is embedded with a programmable algorithm that automatically (or can be patient activated) detects and records asymptomatic cardiac event to include AF. The monitor can be a single or dual channel two-lead system, which allows a total of 30 minutes of recording time and a 300-second prepost event recording for a single event. The data are transmitted transtelephonically and then are posted to a secure web portal within 1 hour of transmission and within 15 minutes for critical transmissions. The Holter monitor (H3+ Holter Monitor) provides 24 to 48 hours of continuous arrhythmia monitoring and recording using a five-lead system. The monitor allows upload to the eCardio laboratory through a secure web transmission and then returns a report to the referring physician within 24 hours. The monitor can also be patient activated for symptoms. Currently, no AF burden report is available, although a new monitor will be released shortly that may be able to deliver an AF burden report. In addition, support through the sending of monitors and filing for insurance is currently not available. The company did not report their algorithm for detection of AF or release their results from the ECG database testing when requested (www.eCardio.com).
NorthEast Monitoring, Inc., Maynard, MA
The NorthEast Monitoring company offers a wide range of products and analysis software. Their Holter monitors (DR180+ digital Holter recorder) can provide up to 14 days of monitoring using only one battery and one memory card (32 MB to 1 GB SD flashcard). Their event monitors (DR200/E “Tel-a-heart”) can provide a two-lead one channel or three-lead two channels of data. The event monitors have the ability to auto detect an event and begin recording it with programmable length periods pre- and postevent requiring no action from the patient; however, the monitor can still be patient activated when needed. They can store up to 90 minutes of data. The company also offers a dual Holter and event recorder (DR200/HE), which can be deployed with a patient in either mode and may be beneficial to maintain sufficient inventory.
The company also provides the software that is necessary to analyze the information obtained from the monitors. Their system uses R to R interval as the primary input for detection of AF. The algorithm uses an exponentially weighted average of the difference in the adjacent RR intervals of all “normal” beats. The weighting factor for both is a function of the difference between the previous four RR intervals at any time and then the ratio of these averages is used to define the detection of AF. However, currently, this company can only provide a practice with all the components to include analysis software necessary to begin a monitoring program but is not set up to provide external support to small practices who do not have the resources to provide a full-scale monitoring program. In addition, an AF burden report is currently not available. The monitor meets the AAMI EC-38 standard for frequency response (www.nemon.com).
CardioNet, Inc, Conshohocken, PA
CardioNet currently offers cardiac event monitoring, Holter monitoring, and mobile cardiac outpatient telemetry (MCOT). Their MCOT system provides beat-to-beat analysis using a three-lead, two-channel ECG, and automatic detection and transmission of clinically significant ECG data through wireless technology. The MCOT system also has the ability to perform remote ECG data fetch and will shortly provide an AF burden report that details any AF >30 seconds. AF is detected using an algorithm that uses RR interval variability and QRS (ventricular activity) morphology but does not directly measure atrial activity; therefore, AF or atrial flutter that does not produce RR variability may not be detected. However, they have recently updated their monitoring to include p wave recognition. For phase 1, the p wave recognition is only used when an irregular irregularity is diagnosed. If an irregular irregularity is diagnosed, then the algorithm goes back to look for p waves. This can still be an issue when measuring AF burden in a patient who is in normal sinus rhythm (NSR) but experiencing premature atrial contractions or another irregular rhythm that can confuse the AF algorithm. Patients can also trigger an analysis if they feel they are experiencing an arrhythmia/palpitations.
The company is also positioned to send patients the monitors, assist patients in hook up, bill patients' respective insurance, provide a return prepaid envelope for the monitor return, send immediate faxed reports on arrhythmias, and directly contact patients for emergent arrhythmias. The company also provides a secure website where report can be uploaded and then retrieved by the physician. They also offer the physician the ability to modify their emergent notification criteria. Patients can also be monitored up to 21 days but may need a battery change; however, the batteries are provided to the patient. In addition, the patient must be in an area with good wireless connections when using the MCOT system.
The company, Cardionet, reports the sensitivity for normal QRS identification using the MIT DB as 99.9% and the positive predictivity as 99.8%. For AF of ≥30 seconds, the reported sensitivity using the MIT DB as 100% and the positive predictivity as 92%. However, with the recent addition of p wave recognition, the sensitivity, and predictivity are 100% for episodes of AF ≥30 seconds based on the MIT and AHA databases (www.cardionet.com).
Biomedical Systems, St. Louis, MO
Biomedical is another supplier of cardiac event and Holter monitors. The event monitors not only use postsymptom and memory loop recorders to capture rhythms but also have auto trigger capability. They offer 24/7 support from a staff of certified technicians and have the ability to assist patients of all ages. Biomedical has web-based system where patients can be enrolled online and provide internet delivery and review of ECG reports. Each monitor shipped to the patient includes batteries, electrodes, and a prepaid envelope for device return; however, each institution is responsible for billing insurance. When pertinent rhythm information is identified or the patient experiences symptoms, the information is transmitted transtelephonically. Their Holter monitors provide 24 or 48 hours of rhythm information using either a 64 Mb or 128 Mb compact flashcard, respectively. The Holter monitor provides three-, five-, or seven-lead three-channel system. There is a 24-hour turnaround time for all Holter reports, which are stored as a PDF files and available online. This company is working with Instromedix to provide a monitor with AF detection capability.
Instromedix, San Diego, CA
Instromedix offers several different types of event monitors to capture arrhythmias that have been diagnosed or to assist in diagnosing difficult to diagnose symptoms that may suggest a cardiac arrhythmia (King of Hearts Express AF recorder, King of Hearts Express + Recorder, and King of Hearts Express Recorder). The King of Hearts Express AF recorder is designed to automatically capture episodes of AF tachycardia and bradycardia. It is a one-channel lead system that requires 2 to 1.5 V “AAA” batteries and can record continuous data usually 7 days but can be extended to 30 days. It has 600 seconds (10 minutes) of memory, which can be programmed for pre- and postactivation time periods. The monitor can capture both symptomatic and asymptomatic events. The asymptomatic events are captured through an AF auto trigger, which is pending patent so that the algorithm for arrhythmia detection was not released. The reports are then transtelephonically transmitted to an office or clinic. Currently, there is no available information about whether the emergency notification criteria can be modified for respective practices. In addition, the monitor does not currently give specific time periods spent in AF. However, the reports are available through an internet service provider.
Their monitor meets the AAMI standards EC 38-94, which means that the monitor will only trigger after the heart rate detector has three valid and consecutive QRS intervals or 10 heart beat arrhythmia to trigger for a programmed bradycardia or tachycardia limits but does not perform analysis. The heart rate detector algorithm was tested against the AHA, MIT, and NST databases but did not report their precise sensitivity and predictivity. They do offer limited external support to practices by shipping the monitors to the patients; however, each practice would be responsible for filing insurance reimbursement (www.biomedsys.com).
MediComp, Melbourne, FL
MediComp provides a monitor that is both a Holter and event monitor (CardioPalR SAVI) in one device. The monitor is programmed to use artificial intelligence along with a patented Holter algorithm for beat identification in combination with a beat-to-beat surveillance to identify AF. The beat surveillance uses four pillars as clinical justification to auto capture. The four pillars are as follows: rate, rhythm, QRS morphology, and p-wave presence. Using these four pillars, the artificial intelligence is constantly reviewing the data sent to the event monitor that is especially important in asymptomatic patients. The event rhythm information is transmitted transtelephonically. The Holter information is obtained once the monitor has been returned to the office. The turnaround time for the Holter and AF report is ∼24 hours. All monitoring is three leads two channels. MediComp will enroll patients through a faxed prescription, send all equipment with a return postage paid envelope, assist patients in hook up, file for insurance, and call patients for problems (www.medicomp.com). Medicomp, Inc. reports the sensitivity for normal QRS identification using the MIT DB as 99.1% and the positive predictivity as 99.91%. They also report the sensitivity for normal QRS morphology using the AHADB for evaluation of ventricular arrhythmia detectors is 99.3% and the positive predictivity of 99.93%. For AF, they report the sensitivity using the MIT DB as 99% and the positive predictivity as 100%. In addition, they report that duration sensitivity using the MIT DB as 93% and the duration positive predictive accuracy as 84%. All monitoring is two leads two channels. In addition, patients can trigger an analysis of their rhythm if they feel they are experiencing an arrhythmia/palpitations (www.medicompinc.com).
The selected criteria used to review the monitors in this report were not meant to be inclusive of all criteria important when selecting a monitor for ones postsurgical ablation monitoring program. However, it was apparent to us when we began our program that these criteria were extremely important to ensure success with our program. Currently, >100 surgical ablations both as a stand alone procedure or performed with other cardiac surgery procedures are performed annually at our institution. The resources that would be allocated for the monitoring program were limited; therefore, it was determined that we needed to work with a company that would provide us full support (ie, billing insurance companies, sending the monitors to patients, assisting patients in hook up, and providing a full report meeting the HRS guidelines at the completion of the monitoring period) to ensure a successful program and an accurate reporting of our return to sinus rhythm rate postsurgical ablation. We, therefore, selected two companies, which we felt met all the criteria noted in this review but most especially the analysis of AF and the provision of AF burden. There was no difference (P = 0.45) in patient compliance between the two monitors. Both monitors have a >70% (72% vs. 78%) compliance for wearing the monitor over the entire monitoring period. In addition to the selection of monitor systems which met our needs, we ensured that all reports would be reviewed by our surgical team and the patient's respective cardiologist/electrophysiologist in determining the final rhythm interpretation as many of the current monitors relay only on R to R variability for interpretation of AF. This is a limitation of many systems because the R to R variability could be caused by other arrhythmias such as sinus arrhythmias or frequent premature atrial contractions. Furthermore, many of the monitor algorithms have not been independently tested to ensure validity of their arrhythmia algorithms; however, several did report their results compared with an ECG database. We are continually evaluating our program to ensure that we are cognizant of what is new on the market and the feedback we receive from our patients.
In addition, we monitor patients for 5 days as we felt that 5 days was an acceptable length of time for patient compliance and to capture AF events as most of the significant events occurred within 72 to 96 hours.10 Further work must be done in determining the most appropriate length of time for monitoring where the most accurate number of events are captured and patients are compliant with the wearing of the monitor. This question may be answered as more reporting is forthcoming investigating the strengths and limitations of an implantable device to include sensitivity and insurance reimbursement compared with the currently Food and Drug Administration-approved external devices. Preliminary results for the implantable device are encouraging, and upon further refinement, such as the amount of time that data can be stored and the ability to provide AF burden, these devices may be another alternative to better understanding the true success of surgical ablation.11
For insurance reimbursement purposes, patients can be monitored up to two times a year if experiencing symptoms or the practitioner is suspicious of recurrence of arrhythmia. However, the cost of each device and what is reimbursed through insurance was not available for this analysis. For our program, we negotiated with our selected monitoring companies to provide service to all our patients regardless of their insurance status/carrier. Patients were responsible only for the co-pay when applicable.
A limitation of this article is that not all the monitors reviewed reported their sensitivity and predictivity. The nonreporting was a result of the respective companies not releasing their AF algorithm or their respective results when tested against the ECG database. Therefore, it is difficult to make a full comparison. But, obviously, the reporting of the arrhythmia algorithm and sensitivity and predictivity should be evaluated before selecting a monitor for one's program.
Therefore, we feel the most important criteria for the selection of a monitor postsurgical ablation includes a monitor whose algorithm captures AF using more than R to R variability, the ability for a patient to trigger a recording when they feel they may be having palpitations, a mechanism which generates a report indicating the amount of time spent in each episode of AF, and the ability to bill for services and follow-up customer service.
Choosing the correct diagnostic method to determine long-term success of the surgical ablation procedure for AF is important and essential. Recent HRS guidelines suggest long-term monitoring for the postoperative maze patient is necessary to determine the success of the procedure. It is therefore essential for a practitioner, before beginning a following ablation monitoring program, to determine what criteria is necessary for the establishment of a successful program. This review sought to compare several available arrhythmia monitors on selected criteria from an established monitoring program to assist the practitioner with choosing a monitor that meets their practice needs and is best for determining the return to sinus rhythm postsurgical ablation especially when working with the asymptomatic patient.
1.Riley M, Marrouche N. Ablation of atrial fibrillation. Curr Probl Cardiol.
2.Calkins H, Brugada J, Packer DL, et al; European Heart Rhythm Association (EHRA); European Cardiac Arrhythmia Scoiety (ECAS); American College of Cardiology (ACC); American Heart Association (AHA); Society of Thoracic Surgeons (STS) HRS/EHRA/ECAS expert consensus statement on catheter and surgical ablation of atrial fibrillation: recommendations for personnel, policy, procedures and follow-up. A report of the Heart Rhythm Society (HRS) Task Force on catheter and surgical ablation of atrial fibrillation. Heart Rhythm
3.Ad N, Henry L, Hunt S, Barnett S, Stone L. The Cox Maze III Procedure Success Rate: Comparison by Electrocardiogram, 24 Hour Holter Monitoring and Long-Term Monitoring. Ann Thorac Surg
4.Zimetbaum PJ, Josephson ME. The evolving role of ambulatory arrhythmia monitoring in general clinical practice. Ann Intern Med.
5.Crawford MH, Berstein SJ, Deedwania PC, et al. ACC/AHA Guidelines for Ambulatory Electrocardiography. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Revise the Guidelines for Ambulatory Electrocardiography). Developed in collaboration with the North America Society for Pacing and Electrophysiology. J Am Coll Cardiol.
6.Rothman SA, Laughlin JC, Seltzer J, et al. The diagnosis of cardiac arrhythmias: a prospective multi-center randomized study comparing mobile cardiac outpatient telemetry versus standard loop event monitoring. J Cardiovasc Electrophysiol.
9.US Food and Drug Administration. Guidance for industry and FDA staff: class II special controls guidance document: arrhythmia detector and alarm. Available at: http://www.fda.gov/cdrh/ode/guidance/1363.html
. Accessed April 1, 2009.
10.Ad N, Henry L, Hunt S, et al. The Cox-Maze III procedure success rate: comparison by electrocardiogram, 24-hour Holter and long-term monitoring. Ann Thorac Surg.
11.Task Force members; EHRA Scientific Documents Committee. Indications for the use of diagnostic implantable and external loop recorders. Europace
This is a timely review article by Drs. Linda Henry and Niv Ad comparing the different arrhythmia monitoring systems that are presently available to assist practitioners in determining the postoperative rhythm of patients after surgical ablation. The consensus statement published in 2007 by the Heart Rhythm Society in conjunction with The Society of Thoracic Surgeons and other international electrophysiological associations identified the need for atrial fibrillation (AF) programs to adopt a consistent method of follow-up and monitoring (Calkins H, Brugada J, Packer DL, et al. HRS/EHRA/ECAS expert consensus statement on catheter and surgical ablation of atrial fibrillation: recommendations for personnel, policy, procedures and follow-up. Heart Rhythm 2007;4:816-861). There is considerable evidence in the literature that the longer the period of monitoring, the more likely one is to discover recurrent atrial tachyarrhythmias. Although the description of the six cardiac rhythm monitoring systems is accurate, there are important limitations to these long-term monitoring systems. All of the present systems use algorithms to detect recurrent AF that are far from fool-proof. These algorithms look at R to R variability that can be caused by arrhythmias other than AF such as sinus arrhythmia and frequent premature atrial or ventricular contractions. Generally, the monitoring algorithms have not been independently tested in terms of their sensitivity and specificity. If you use these systems, it is necessary to review the recorded electrocardiograms to ensure that the rhythm is actually an atrial tachyarrhythmia and not a false-positive. It is not adequate to simply read the reports because decisions on whether to discontinue anticoagulation have important repercussions for the patient. The shortcomings of these monitoring systems have led our group to abandon them in favor of 24- and 48-hour Holter monitoring. At our institution, these Holter electrocardiograms are reviewed by board-certified cardiologists, thus ensuring greater accuracy in diagnosis and obviating the need for busy surgeons to have to review numerous rhythm strips. Another benefit is the improved patient compliance with the shorter periods of monitoring. When you follow hundreds of outpatients, this saves considerable time while still meeting the present guidelines (as defined by the recent consensus statement) of at least a 24-hour continuous monitoring.
© 2010 Lippincott Williams & Wilkins, Inc.