Cardiac Surveillance at Home

Outpatient ECG Each of these modalities has different characteristics that influence their use. For example, Holters are inconvenient for recording over about 48 hours and the diagnostic data can be delayed by up to several days, while event recorders can capture only symptomatic events.1 Telemetry provides ECG monitoring with rapid response capability and is routinely used as a potentially life-saving measure for essentially 100% of the nation s heart attack victims who are monitored during their hospital stay,2 but is not available on an outpatient basis. Cardiac Surveillance As the nature of health care delivery in the United States changes, one consistent theme is that hospitals may soon be limited to very sick individuals, thus, more will be done on an outpatient basis and at alternative lower cost sites. Recently, a new monitoring modality was developed and became routinely available, combining many of the advantages of previous technology. It provides real-time ECG arrhythmia monitoring (Cardiac Surveillance) for any potential expanded patient sites, taking it all the way to the patient s home. The system is known as the HEARTLink II, a fully automatic, real-time cardiac arrhythmia detection and alarm system that has specifically been designed to operate in a home environment. It is not intended for individuals who are at high risk of developing lethal arrhythmias such as primary ventricular fibrillation or sustained ventricular tachycardia, but rather for those who might benefit from timely identification of less malignant arrhythmias. HEARTLink basically operates as telemetry for the home, providing cardiac surveillance with automatic rapid notification of rhythm abnormalities. Illustrating its compatibly with in-hospital telemetry, an essentially identical system, HEARTtrac, has been cleared by the FDA for use in hospital step-down units. Monitoring Modes Operationally, the system monitors and labels electrocardiogram patterns, automatically detects any rhythm abnormality based on 16 different proprietary diagnostic algorithms, and then automatically transmits them via a standard telephone to a central monitoring station. It operates without patient intervention and has four operational modes: real-time automatic event mode, real-time display mode, patient-activated mode for symptomatic events, and patient help request. The device is most commonly left in automatic mode. Any automatically transmitted rhythm strip is seen within 15 seconds by trained operators at a central location that is manned 24 hours/day. As occurrences of arrhythmias are detected, transmitted and displayed at the central station, they are viewed by medical professionals and handled per protocol, stored and later merged with trend data developed at the patient site. This allows a Holter Scan equivalent report to be prepared for any past 24-hour period that the patient is on the system providing a parallel to conventional Holter equipment. Should a patient feel a symptom, a push button is provided, which immediately sends a strip establishing compatibility with traditional event recorders. The same push button is used to activate a Help Request for use in a possible non-cardiac emergency. Additionally, during the real-time display mode, medical technicians can log on from a remote terminal and display and view ECG patterns in real time as they occur. Monitoring Service HEARTLink II is available to physicians as a service called Telemetry @ Home, initially targeted for patients who can benefit from its ability to: 1) provide timely diagnosis and notification of arrhythmias, 2) detect asymptomatic arrhythmias, 3) be conveniently worn for a period longer that a conventional Holter, or 4) operate without patient intervention. A patient transmitter (Patient Module; Figure 1) digitally transmits cardiac information to a ruggedized computer box (Tele-Link; Figure 2) that monitors, analyzes, stores and quickly transmits predetermined cardiac events to a Surveillance Lab (Independent Diagnostic Testing Facility, or IDTF). This analysis is real-time and the analysis/transmission time to the Surveillance Lab is typically less than 12 seconds. Detection parameters can be changed per physician instructions from the Surveillance Lab. The patient s telephone line works as a normal telephone line when the Tele-Link processor is not transmitting data. If the patient is on the phone when a transmission is initiated, the voice call is disconnected by the Tele-Link, allowing the data call to proceed. Extensive error detection provisions are made for most situations that may arise outside of the surveillance itself (e.g., low battery, leads off, patient out of range, phone line down, etc.) and a four-hour battery backup is included in the Tele-Link Processor itself. Patient Enrollment During patient enrollment, the physician determines arrhythmia detection parameters and notification criteria for the patient. Detection criteria are programmed into the Tele-Link during the patient set-up process to accommodate for the fact that each patient situation is unique. Once Cardiac Telecom receives this information, verification is immediately faxed back to the physician, with dates of the patient surveillance. The physician has the opportunity to modify any of the default diagnostic criteria prior to the patient coming on service (and again, has the ability to change these during the service as well). As enrollment proceeds, the lab contacts the patient, explains the process of surveillance to them, and, for the entire surveillance period, speaks with each patient at least twice a day as part of normal protocol. To ensure patient compliance, the system is extremely easy to use, as all the patient needs to do is wear the transmitting device and then change the battery when needed. Most patients are even capable of connecting the equipment up in their home, and for those who cannot, arrangements can be made to have it done for them. Although the system performs significant cardiac surveillance and analysis on a continuous basis, this is done transparently. Usually, the lab will call each patient s Tele-Link every evening and download additional information about the patient s cardiac activity. Daily diagnostic reports are then sent to each patient s physician for review. Although clinicians normally associate telemetry with immediate on-site care, this is not the case with HEARTLink II. Very high-risk patients should be avoided; however, the lab is staffed with highly trained EMT-Paramedics and/or RNs 24 hours a day, 7 days a week. Based on the acuity of the patient, local EMS facilities may be notified upon patient hook-up that a patient is on the system for cardiac surveillance, and an explanation of HEARTLink II is given at that time. Regardless of patient acuity, the lab obtains information about each patient s local EMS facility prior to patient hook-up. Clinical Results Clinical data have been accumulated to characterize utility of the Telemetry @ Home system. In the Fall of 1997, a multi-center clinical trial was instituted by the Cardiothoracic Service Line at the Department of Thoracic and Cardiovascular Surgery, St Francis Hospital; the Cardiac Arrhythmia Center at the Department of Cardiovascular Disease, Washington Hospital Center; and at Cardiac Telecom Corporation. The two center s IRBs and the Food and Drug Administration (FDA) approved a clinical trial to compare the usability of the HEARTLink II Telemetry @ Home Monitoring System to standard event recorders. Patient outcomes were also tracked at one site, St. Francis Hospital, and are reported first, while the usability data is presented under the section Comparison to Event Recorders. During a seven-month period from November of 1997 through June 1998, a total of sixteen patients successfully completed the clinical trial. Sixteen patients were discharged to home on either an event recorder or the HEARTLink II device, and then subsequently switched to the other device. Patients were randomized utilizing the last digit of their social security number to determine what device would be used first. A total of 100% of the patient population on the Cardiothoracic Service Line has their outcomes reviewed, analyzed and measured via a Variance Monitoring Methodology (VMM). Variance Monitoring measures demographic data, resource utilization and outcome measures for all patients undergoing surgical interventions. Currently, this monitoring methodology separates patients into major clinical pathway categories such as patients undergoing open-heart surgery, thoracic surgery procedures, major vascular and peripheral vascular procedures. The historical data gathered by VMM can be used as a control group to compare outcomes of patients using HEARTLink II and those who did not. Table 1 compares demographic and outcome information for the HEARTLink II patient population and the general Variance Monitoring Data (VMD) patient population. The first tables indicate the two groups being approximately the same age, with the HEARTLink II patients being only slightly older. The mean number of grafts per case for those patients on the HEARTLink II study was 3.73 (Table 2). The mean number of grafts for each of the patients during the clinical trial for the VMD patient population graft/patient ranged from 3.05-3.72. Graft numbers are similar in the two populations of patients. The data in Table 3 represent the mean length of stay for patients undergoing open-heart surgery. Unlike the HEARTLink II Patient Protocol, where only coronary artery bypass patients were included, the VMD includes patients undergoing coronary artery bypass procedures, valve procedures, combined procedures, emergent procedures, redo procedures, etc. The VMD does exclude outlier patients (defined as LOS > 32 days). The length of stay for the HEARTLink II patients was between 2 to 3 days less than the patients tracked by VMD. The fifth post-operative day (5POD) represents the targeted day of discharge for the Cardiothoracic Service Line Open Heart Surgery Clinical Pathway patients (Table 4). The overall post-operative LOS for the HEARTLink II patients was 4.87 days, which represents a 100% compliance with the Clinical Pathway. The range of LOS was discharge on the third post-operative day to as late at the seventh post-operative day of discharge. The rates of compliance per month for all patients monitored through VMD are as shown in the VMD data table, but all were less than 100%. Comparison to Looping Event Recorders Along with the clinical outcome data, during the multi-center trial, data to support a 510(k) submission for the HEARTLink II system and to compare it to currently marketed cardiac event recorders were also gathered. These studies were to test the effect of home environmental factors on the HEARTLink II s ability to monitor patients heart signal data and the utilization of commercial communications networks to transmit this data to a central monitoring system. The accuracy of transmitted data (whether or not the event was transmitted and received correctly), had been previously verified in laboratory testing and in the hospital environment. Two areas of comparison were to be evaluated. The first area tested the usability of the HEARTLink II system. The second area determined the equivalence of the HEARTLink II system s and the event recorder s transmitted data. These areas formulated two hypotheses that would be tested in both clinical and healthy volunteer trials. To evaluate patient usability, the same patients as in the clinical study were utilized. Each patient was randomly assigned to one of two groups. The first group consisted of patients who wore the HEARTLink II system at the start of the study (Trial 1). The second group consisted of patients that wore the event recorder at the start of the study (Trial 1). After three days, the patients were crossed over to the alternate monitoring system for an additional three days (Trial 2). At the end of each three-day segment, it was noted if the patient was able to complete that particular segment of the study. A total of 19 patients were to be remotely monitored in their home environments. The outcome to be measured was the number of patients that complete each segment of the study with either the HEARTLink II system or the event recorder. Each study segment, for every patient, is represented by either a Yes or a No. Of the 19 patients that were included in the data analysis, there were 89.5% (17 patients) who were able to complete the study with the HEARTLink II system. One patient (Patient #9) was unable to complete the HEARTLink II system study segment because the patient s home was located in an area that required the hospital s Central Station to dial a newly assigned area code (724 instead of 412) to complete the long distance phone call. This fact had not been determined until the patient was disconnected from the device and the study segment classified as a No (unable to complete study). The HEARTLink II system, however, responded with an appropriate error message that the patient was not being monitored. To correct future occurrences of this operator error, all hospital personnel were trained on how to successfully detect and correct this situation. Another patient, who was initially connected to the HEARTLink II, developed atrial fibrillation after less than 1 hour and was subsequently hospitalized. The HEARTLink II System correctly detected this arrhythmia from this patient and transmitted the alarm and the ECG data to the Central Station. Arrangements were made to have this patient treated at a local hospital after the patient s physician was notified of this arrhythmia. Although the device functioned properly, the patient did not complete the study, and the result was coded as a No. Of the 19 included patients, there were 78.9% (15 patients) who were able to complete the study with the event recorder system. One patient (Patient #3) was sent home with a defective event recorder. Another patient (Patient #10) was unsuccessful at downloading the data from the event recorder over the phone, in spite of the verbal instructions given to the patient by the Central Monitoring station staff. Only a total of three events were properly initiated and transmitted by this patient to the event recorder Central Monitoring station. The third patient (Patient #15) appeared to be unable to understand the concept of the operation and download procedures of the event recorder. This patient s device was also suspected of having an intermittent problem that affected the patient s ability to utilize it. The fourth patient never reached the event recorder phase, having been hospitalized. These data suggest that the success rate using an event recorder and HEARTLink II systems does not significantly differ. The number of patients who completed the study segment using an event recorder is comparable to the number of patients who completed the study segment using HEARTLink II. Successful completion of the study with either the event recorder or HEARTLink II systems required the patient to perform all tasks associated with the maintenance of recording electrodes, battery changes, activation of events and in the case of the event recorder, successfully but manually downloading the data to the central monitoring facility over the telephone. Patients using the HEARTLink II system were equally as successful at the performance of all these associated tasks as with the event recorder system. Cardiac Telecom also performs an event recorder service and has compared data (Table 5) regarding symptomatic and asymptomatic transmissions from event recorders to Telemetry @ Home. In all these cases, Cardiac Telecom has actually conducted the service. The results of this tabulation, based on 59 patients, show that the vast majority of transmissions made by the patients (117 out of a total 177) were NSR, and yet these are symptomatic in the sense that the patient felt something before initiating the transmission. In addition, we have pulled a subset of patients that were on Cardiac Telecom s Telemetry @ Home service this year and analyzed the results of the transmissions received from those on-service patients. It is interesting to note, that in contrast to the results of transmissions received from event recorders, the vast majority of the true events detected and transmitted to the lab were asymptomatic (greater than 95%). That is, the events that triggered a transmission to our lab at the time of transmission were not felt or reported as symptomatic by the patients. Note that the patients themselves were not asymptomatic, e.g., they were on service for a specific complaint such as syncope, but rather the events were not felt by the patients. Therefore, it is believed that none of these transmitted events would have been transmitted to a lab had the patient been using an event recorder. Furthermore, it is believed that these asymptomatic events contributed to the patient being symptomatic. Several patient alarms were transmitted to the lab in this subset, when the patient believed they were experiencing cardiac symptoms. In every case, those specific patient-activated transmissions on our Telemetry @ Home service showed the patient in normal sinus rhythm. Discussion The clinical data suggest that the hospital length of stay may be reduced in post-bypass patients with the HEARTLink II system, and that the desired clinical pathway for that patient is more likely to be met. This may be due to the attending physician s increased confidence that post-discharge patient problems are more likely to be identified and dealt with quickly, as illustrated by the following clinical vignette: Patient #1, a 66-year-old male, underwent a coronary artery bypass times three in November of 1997. He was discharged to home on his fifth post-operative day, which complied with the Clinical Pathway goals. His post-operative course was uneventful. He was discharged to home on the HEARTLink II device. Routine review of trended HEARTLink II strips indicated an excessive number of PACs. Viewed by themselves, they did not warrant an alarm situation; however, when an eight-hour trend was reviewed, it was believed that the patient was on the verge of converting to atrial fibrillation. The cardiologist was contacted, and oral anti-arrhythmic medications were initiated. The patient was monitored 48 additional hours, during which time the PACs stopped. Both the nurse and cardiologist believed this patient would have converted to atrial fibrillation requiring a higher level of intervention if the preventive action had not occurred. Patients should find it at least as easy if not easier to use than an event recorder. For example, some patients, particularly several with dementia, are unable to use conventional event recorders because of the complexity involved, but can use Telemetry @ Home since it is fully automatic. Another patient, a deaf mute, was also unable to use an event recorder since usage involves responding to beeps and talking on the phone. Additionally, there may be value in detecting asymptomatic events in otherwise symptomatic patients. ECG rhythm abnormalities may not correlate well with actual perception of symptoms, resulting in many event recorder transmissions of normal sinus rhythm, at least for those patients placed on an event recorder service at Cardiac Telecom. In contrast, most HEARTLink II transmissions contain an arrhythmia. Very few transmissions are initiated by the patient, and are therefore asymptomatic. A difficult to obtain symptom-rhythm correlation has been reported by others during spontaneous syncope.3 Table 6 lists the differences among the various technologies, HEARTLink II, Holter monitors and looping event recorders, and should aid the clinician in choosing a suitable modality for his patient. Uses Telemetry @ Home has been very successfully used with post-surgical coronary artery bypass graft patients who have experienced post-procedure periods of atrial fibrillation, often asymptomatic but potentially harmful. Timely identification allows interventions preventing greatly increasing risk of stroke. Patients are usually on the system from 5 to 10 days, and though they must stay home to be monitored, this is not a serious inconvenience for these patients. The patients may leave the home from time to time for short periods (e.g., for a doctor s appointment), but will not be monitored once they go beyond a maximum 500-foot range. The system has also been used very successfully with difficult to diagnose arrhythmias or when the patient is undergoing either drug titration at home or there has been a change in medication with potential pro-arrhythmic side effects where the nearest competing technologies, event recorders and Holter monitors could not be used or have been used with negative yields. Many patients have had failed Holters, but arrhythmias were still suspected, which Telemetry @ Home often quickly identified. These patients clearly benefited from the advantages of the system. An additional advantage is the system s ability to identify artifact on call-in, allowing the operator to advise the patient to re-apply electrodes. On a conventional Holter, it is possible to record an entire tape of artifact should there be an electrode problem, the problem being identified only after the fact while the tape is being scanned. Telemetry @ Home gives the physician the data required, when it occurs (real-time) to allow for timely intervention. Some case study data suggest that Telemetry @ Home is keeping patients out of the emergency room and is helping to reduce readmissions for lengthy stays, which also accounts for significant cost savings. The service is also proving efficacious in assisting physicians to determine quickly the need for pacemakers where that device is indicated. The system has also helped intervene very quickly when medical attention has been required and has assisted in saving several patient lives. It is of interest to note that several patients have only been on service for a matter of hours when a timely and potentially life-saving intervention has been facilitated or initiated by the Telemetry @ Home service. * at the time of the research described in this paper, now with Jefferson Regional Medical Center, Pittsburgh, PA