Monitoring Methods

Monitoring of AT/AF has traditionally relied on patient-reported symptoms or the intermittent use of external devices to capture asymptomatic episodes. However, both approaches have significant limitations in the diagnosis and management of atrial arrhythmias.

Symptoms Versus External Recorders

Monitoring AT/AF on the basis of patient symptoms is appealing because it is relatively inexpensive and has clinical relevance. However, studies have shown that the vast majority of AT/AF episodes are asymptomatic and that most symptoms attributed to AT/AF are not actually associated with arrhythmia.^2,3,8,9 Therefore, even if supplemental event recorders are used to verify the presence of an arrhythmia during symptoms, most atrial arrhythmias will escape detection because of lack of symptoms. Since the risk of stroke is similar among patients with symptomatic and asymptomatic AT/AF, current anticoagulation guidelines do not differentiate on the basis of symptoms,¹⁰ which emphasizes the importance of identifying asymptomatic AT/AF.^10–12

Because of the highly insensitive and nonspecific nature of patient symptoms, external devices were developed to monitor asymptomatic episodes of AT/AF. External devices such as Holter monitors are commonly used to continuously record cardiac data for 24 or 48 hours. Newer systems, such as mobile cardiac outpatient telemetry, are capable of monitoring patients continuously for up to 30 days and have been shown to increase the yield for arrhythmia detection compared with a single 24-hour Holter monitor.^13,14 These systems have the benefits of moderate cost and being a noninvasive procedure. However, these external devices often are bulky and interfere with daily activities such as showering. In addition, the patch electrodes can cause skin irritation with prolonged use. Consequently, patient compliance with such systems often is quite low.^8,13 Furthermore, since external monitoring can only be performed intermittently over time and for relatively brief durations, the likelihood of missing paroxysmal episodes of asymptomatic AT/AF episodes is quite high.

Implantable Monitors

Subcutaneous Monitors

One strategy to address the need for continuous monitoring in a broader population of patients with AF is to use small implantable devices that are capable of monitoring atrial arrhythmias continuously over the lifetime of the device with a high sensitivity and a high specificity for AT/AF detection.¹⁷ These devices, about the size of a cigarette lighter, do not have intracardiac leads, as the bipolar recording electrodes are generally located on the surface of the device itself. These monitoring devices are placed subcutaneously, are relatively small in size (3 to 5 cm), and are typically implanted in the left pectoral region in the precordial area, though axillary implants have been described (Figure 89-1, A). Implantation parallel to and in an intercostal space is most desirable for avoidance of local erosion and aesthetics. A local anesthetic agent is infiltrated at the site of entry and in the adjoining subcutaneous tissue where the device pocket will be made. Typically, a small (1 to 2 cm) vertical or horizontal incision is made at one end of the pocket, and the pocket created by blunt dissection (Figure 89-1, B). The device is placed and the electrogram quality assessed using a hand-held wand or wireless connection by the device programmer. If clear electrogram quality and satisfactory amplitude are obtained, the pocket can be closed. If not, device repositioning is needed so that the surface electrodes have an appropriate vector to obtain quality recordings. Although a small skin incision is required to implant these devices, patient compliance is not an issue, as these devices rarely interfere with normal daily activities. Until recently, subcutaneous monitoring devices had only been available for the diagnosis of syncope, but they have now become available for the monitoring of atrial arrhythmias as well. Patients are able to record the occurrence of symptoms in the memory of the device with an external activator, and these data can be retrieved later by the clinician.

FIGURE 89-1 Surgical placement of an implantable loop recorder device. A, Schematic of potential locations in the left-sided intercostal spaces of the recorder. B, The recorder being inserted subcutaneously through a 2-cm incision with local anesthesia. Longevity of the recorder is 3 years. Event storage with electrocardiogram is 49.5 minutes: 22.5 minutes for patient-activated episodes and 27 minutes for automatic activated episodes (ventricular tachycardia, fast ventricular tachycardia, asystole, bradycardia [1-minute episodes], atrial tachycardia, and atrial fibrillation [2-minute episodes]). Total event logs include 180 automatic and 10 patient activated.

Comparison of Monitoring Methods

Methods of identifying patients with AT/AF through their symptoms, intermittent external monitoring, and continuous monitoring with implantable devices have been compared in a recent study.¹⁸ Symptom-based and intermittent external monitoring methods were shown to have significantly lower sensitivity (range, 31% to 71%) and negative predictive value (range, 21% to 39%) for identification of patients with AT/AF (Figure 89-2) and underestimated AT/AF burden, compared with continuous monitoring. The reported efficacy of clinical procedures such as pulmonary vein ablation for the treatment of AT/AF can vary greatly, depending on how the arrhythmia is monitored. One study reported a success rate of 70% on the basis of symptoms alone and only 50% when intermittent external monitoring was also considered.⁸ These results were achieved despite the fact that 47% of the patients did not complete the external monitoring protocol and compliance with the monitoring schedule was only 42%. Such results highlight the clinical need for continuous monitoring that does not rely on patient symptoms or patient compliance.

FIGURE 89-2 Sensitivity and negative predictive value for identification of patients with any atrial tachycardia/atrial fibrillation episodes identified by various intermittent external monitoring methods or patient symptoms. Annual, quarterly, and monthly refer to 24-hour Holter recordings.

Data Collection and Organization

While implantable devices are capable of continuously monitoring cardiac rhythms over the life of the device, two important reasons why this information must be condensed and summarized must be considered. First, memory limitations of the device would not permit continuous storage of electrogram and other data for each cardiac cycle. A patient experiences approximately 100,000 heartbeats per day and 6 months may elapse between follow-up visits, at which time the data can be extracted from the device via telemetry. This means that the device might monitor more than 18 million heartbeats between clinic visits. Second, this amount of data would be far too much for the physician to process and interpret for each patient. The challenge for device manufacturers is to convert this vast amount of detection information into diagnostic information that is manageable for both the device and the clinician.

To accomplish this, the device stores information on atrial arrhythmias in a hierarchical fashion. Extremely detailed information is stored for only a small subset of the episodes, while very general information is tabulated across all episodes. For example, one of the most memory-intensive pieces of information is the electrogram waveform. These data can be crucial for the clinician to verify that the device is detecting correctly and aid in troubleshooting when it is not. Because this information requires extensive amounts of memory (as well as additional battery resources), only small portions of electrograms from a select number of episodes are stored by the device. On the other end of the spectrum, a continuous running tally of the duration of all AT/AF episodes is important so that the physician can assess the AT/AF burden experienced by the patient. Between these two extremes are a variety of other parameters that may be tabulated per episode, per day, or per follow-up period. The goal is to strike a balance between not providing enough information to be clinically useful in managing the patient and providing too much data that would overwhelm the clinician. Specific examples of device diagnostics for rhythm control, rate control, and anticoagulation management of patients with AT/AF will be presented in the remainder of this section.

Pulmonary Vein Ablation

Figure 89-5 shows a series of patients who were treated for AT/AF by undergoing a pulmonary vein ablation procedure anywhere from 6 to 36 months after device implantation. These patients had a pacemaker implanted for their bradycardia, and the device was subsequently used to monitor the effectiveness of the ablation procedure. Patient data are aligned such that the date of ablation corresponds with month 0. The amount of AT/AF experienced each day is plotted over the follow-up period and shown in blue. Despite an improvement in symptoms and quality of life in all patients after the ablation, many still experienced significant amounts of AT/AF.¹⁹ This example also illustrates that after an ablation procedure, patients can go for many months without an AT/AF episode and subsequently experience recurrences that would be difficult to record with intermittent monitoring methods. This long-term trending information enabled the physicians conducting the study to perform additional ablation procedures (indicated by the red arrows) in select patients who did not respond completely to the initial procedure.

FIGURE 89-5 Daily atrial tachycardia/atrial fibrillation (AT/AF) burden both before (PRE) and after (POST) pulmonary vein ablation in 14 patients. Gray shading indicates the follow-up experience for each patient, and the blue lines indicate the AT/AF burden experienced each day. Repeat ablation procedures are indicated by the arrows.

Antiarrhythmic Drugs and Cardiac Pacing

Response to antiarrhythmic drug therapy can be assessed by device-based diagnostic data. Figure 89-6 shows an example of change in AT/AF burden, with conversion of persistent AF to paroxysmal AF, after the initiation of sotalol. The patient continued to have symptomatic paroxysmal AF and was then transitioned to dronedarone, which resulted in the resolution of symptomatic paroxysmal AF.

FIGURE 89-6 Example of an implantable pacemaker report showing monitoring of atrial fibrillation (AF) burden, episodes, and pacing interventions over a 1-year period. The relationship to drug therapy is shown. The patient was in persistent AF during propafenone therapy, and introduction of sotalol converted this to paroxysmal AF. The patient remained symptomatic and was switched to dronedarone, which resulted in improvement in symptoms and reduction in events. AT, Atrial tachycardia; CV, cardiovascular.

Recurrence Patterns

While there is clinical value in knowing whether or not a patient has experienced AT/AF and the percentage of time that has been spent in AT/AF, understanding the pattern of these AT/AF recurrences may be of additional value. The patient data in Figure 89-7 show a very consistent recurrence pattern that persists for the entire 14-month period covered by device diagnostics. On closer examination, it was found that this patient experienced AT/AF predominantly on weekends, with episodes generally initiating on Fridays or Saturdays and typically terminating on Sundays or Mondays. With intermittent arrhythmia monitoring, it might have been erroneously concluded that this patient was either always in AT/AF or always in sinus rhythm, depending on the day(s) selected for monitoring. With continuous monitoring, a more complete picture of the patient’s arrhythmia can be obtained, which may provide additional insight into the patient’s disease and lead to different treatment strategies. In addition to providing more complete information on individual patients, device diagnostics can inform us on the nature of atrial arrhythmias across entire populations. Figure 89-8 shows the percentage of patients with implantable devices (pulse generator, ICD, and CRT) who experienced at least 5 minutes of AT/AF over each day of the year.²⁰ A clear pattern is seen, with a peak incidence in the month of May and a minimum incidence 6 months later in November. While the underlying cause of this cyclic variation is not fully understood, it is interesting to note that the occurrence of stroke has been shown to exhibit a similar recurrence pattern.²¹

FIGURE 89-7 Unique and repetitious atrial tachycardia/atrial fibrillation (AT/AF) burden pattern from a patient who predominately experienced AT/AF on Saturdays and Sundays. CV, cardiovascular.

FIGURE 89-8 The percentage of northern hemisphere patients who experience at least 5 minutes of atrial tachycardia/atrial fibrillation (AF) on any given day, with a peak in May and a valley in November.

Transition from Paroxysmal Atrial Fibrillation to Persistent Atrial Fibrillation