Brief Report

electrocardiographic monitoring in the hospitalized patient: a diagnostic intervention of uncertain clinical impact

Todd S. Larson, William J. Brady MD?

Department of Emergency Medicine, University of Virginia, Charlottesville, VA 22901, USA

Received 16 November 2007; revised 5 December 2007; accepted 5 December 2007

Abstract Electrocardiographic monitoring, also known as telemetry monitoring or cardiac monitoring, focuses on the detection of clinically significant dysrhythmia. Although electrocardiographic monitoring has advantages and is undoubtedly invaluable in certain patients, significant overuse of cardiac telemetry monitoring does occur. The impact of this inappropriate use increases the cost of healthcare and can delay the admission process, an unpleasant burden which the hospital and its patients must bear. In addition, the liberal use of monitoring in unnecessary situations may give the hospital staff a false sense of security and/or desensitize them to alarms. In many instances, electrocardiographic monitoring may not be necessary. This article will review the literature regarding inpatient telemetry and its impact; furthermore, we will suggest high-yield criteria for its application among the inpatient population.

(C) 2008

Introduction

Many patients admitted to the hospital with cardior- espiratory and other acute syndromes are placed on electrocardiographic monitoring. Such electrocardiographic surveillance theoretically allows the clinical staff to monitor admitted patients for the development of dysrhythmia, both brady- and tachydysrhythmia-in essence, identify patient decompensation and/or clinical deterioration as it occurs rather than after the event. Electrocardiographic monitoring, also known as telemetry monitoring or cardiac monitoring, focuses on the detection of clinically significant dysrhyth- mia, rather than the diagnosis of acute coronary syndrome (ACS) or acute pulmonary edema.

Although electrocardiographic monitoring (ECGM) has advantages and is undoubtedly invaluable in certain patients, significant overuse of cardiac telemetry monitoring does occur. The impact of this inappropriate use increases the cost

* Corresponding author.

E-mail address: [email protected] (W.J. Brady).

of healthcare and can delay the admission process, an unpleasant burden which the hospital and its patients must bear. In addition, the liberal use of monitoring in unnecessary situations may give the hospital staff a false sense of security and/or desensitize them to alarms [1,2]. In many instances, electrocardiographic monitoring may not be necessary. This article will review the literature regarding inpatient telemetry and its impact; furthermore, we will suggest high-yield criteria for its application among the inpatient population.

Current state of practice: electrocardio- graphic monitoring

Currently, practitioners use electrocardiographic monitor- ing for patients admitted with various cardiovascular ailments as well as a range of other clinical entities, including neurologic, respiratory, traumatic, metabolic, and toxicologic syndromes. Furthermore, among these various syndromes, the degree of clinical risk varies from significant to negligible. As noted, electrocardiographic monitoring

0735-6757/$ - see front matter (C) 2008 doi:10.1016/j.ajem.2007.12.003

targets the patient at risk for the development of clinically significant dysrhythmia.

Typically, emergency physicians admit patients with potential ACS to ECGM units [3,4]. This has been a standard practice for decades because patients experiencing an acute myocardial infarction have an increased risk of life- threatening arrhythmias [5,6]. Fifty years ago, Intensive care units were the primary means of managing such patients. Later, the use of intermediate-level telemetry wards became popularized for lower-risk patients owing to the prohibitive financial cost of ICUs [7,8]. Currently, the American Heart Association endorses the use of ECGM for patients with chest pain who may be experiencing cardiac ischemia [9]. Recent reviews of this technology, however, are questioning this widespread use of ECGM for less concerning patients with chest pain, with many studies demonstrating a lack of utility for monitoring individuals determined to be low risk by varying criteria.

Limited utility of electrocardiographic monitoring for low-risk patients

Currently, the typical practice in most North American hospitals is widespread application of ECGM for patients admitted for the management of chest pain, dysrhythmia, syncope, ingestion, etc. And yet numerous studies have been performed which demonstrate the limited utility of such monitoring for these admitted patients, many of whom do not meet intermediate- or high-risk criteria. For instance, a prospective cohort study completed by Hollander et al [10] found that dysrhythmias requiring physician intervention during hospital admission are unlikely to occur in patients with chest pain with low-risk electrocardiographic patterns. The study consisted of patients with nontraumatic chest pain who presented to the ED with normal or nondiagnostic 12- lead Electrocardiograms and were admitted to non- ICU ECGM beds for rule-out myocardial infarction includ- ing chest radiography, serial biomarkers, and repeat ECGs; these investigators followed the patient’s hospital course, reviewing the impact of cardiac monitoring on patient management and outcome. A total of 460 patients were admitted to telemetry beds during the study. Only 4 (0.9%) patients had a notable cardiac arrhythmia-1 supraventri- cular tachycardia, 1 sustained ventricular tachycardia, and 2 bradyarrhythmias. It should be noted that the patient who had an episode of sustained ventricular tachycardia experienced this event after an aortic dissection repair. Besides this one episode of ventricular tachycardia after surgery, no other patients underwent Treatment intervention due to ECGM- derived clinical data. Of the 261 patients who presented with a normal or nonspecific ECG, no one benefited from continuous cardiac monitoring.

Several other authors have reported similar results. Snider et al [11] demonstrated that low-risk patients rarely benefited

from electrocardiographic monitoring during their inpatient stay; conversely, these investigators noted that the high-risk subset of patients did demonstrate potential benefit. These investigators performed a prospective cohort study of 414 consecutive ED patients who were admitted to a non-ICU telemetry unit for rule-out ACS; importantly, these investi- gators considered this population from different levels of cardiovascular risk-low vs high risk. One hundred one patients (24.4%) were considered low-risk based upon atypical chest discomfort and normal 12-lead ECG; none of these patients demonstrated clinically significant abnorm- ality on ECGM during hospitalization. At the opposite end of the risk spectrum, a number of high-risk patients experienced significant arrhythmic events during the admission. One hundred 7 patients (28.5%) experienced at least one cardiac event on telemetry with 47 (11%) of these patients under- going an alteration in management; the alterations in management in this subgroup included addition of antiar- rhythmic medication (5.6%), medication dosage change (2.9%), electrophysiology study performance (1.9%), ICU transfer (1.9%), and elective electrical cardioversion (0.5%), among other less frequent manipulations.

Another investigation performed by Saleem et al [12] reviewed 105 consecutive patients who presented to the ED with Low-risk chest pain and were subsequently admitted to a cardiac monitored unit. The authors defined this low-risk category in the ED as follows: single negative biomarker, an absence of electrocardiographic abnormalities (ST segment and T wave), hemodynamic and rhythm stability, and no requirement for invasive intervention-importantly, these criteria are all ED-derived clinical data. The authors found that telemetry did not alter the management of any patient in their study nor did this intervention identify individuals at increased risk of long-term poor outcome which occurred over approximately 4 years. Both Snider et al [11] and Saleem et al [12] found that a simple risk stratification performed in the ED could successfully identify a group of patients who were unlikely to benefit from Continuous electrocardiographic Monitoring during hospital admission.

Although some studies focus on alteration of management as a proxy measure for the utility of ECGM, Schull and Redelmeier [13] actually considered the number of lives saved by electrocardiographic monitoring in their analysis. The authors conducted a retrospective study of ECGM unit admissions with attention paid to cardiac arrest and ultimate outcome in a tertiary care center to examine the usefulness of cardiac monitoring on inpatient Cardiac arrest survival. During the 5-year study period, a total of 8932 cardiac arrest patients were admitted to the unit. Twenty patients (0.2%) experienced cardiac arrest while on the telemetry ward. Unfortunately, ECGM identified the clinical deteriora- tion at the time of occurrence in only half of these individuals. Three of the 20 cardiac arrest patients survived and were discharged home; the arrest onset was identified by telemetry in 2 of the 3 survivors, resulting in a monitor- signaled survival rate among telemetry unit patients of a

mere 0.02% among almost 9000 patients. Interestingly, in this paper, only 50% of patients demonstrated the cardiac arrest via electrocardiographic monitoring-indeed, the widespread application of electrocardiographic monitoring did not produce a substantial Survival benefit, even in those patients who experienced cardiac arrest.

All too often, clinicians overestimate the usefulness of cardiac monitoring. For example, a prospective cohort study completed by Estrada et al [14], which examined 2240 patients admitted to a telemetry unit, emphasized the separation between perception and reality. Clinical indica- tions supporting the admission to the ECGM unit included chest pain (55%), arrhythmias (14%), Acute congestive heart failure (12%), and syncope (10%). Electrocardiographic monitoring led to an alteration in management in 156 (7%) patients; in addition, electrocardiographic monitoring was found to be useful but did not result in a Change in management in another 127 (5.7%) patients. A total of 241 patients in the study were eventually transferred to an ICU from the telemetry ward (10.8%); only 19 (0.8%) of these patients, however, were transferred to the ICU based on data (ie, arrhythmia) provided by ECGM. In this study, physicians viewed telemetry to be useful in 51% of patients admitted with arrhythmias, in 16% of patients admitted after a syncopal episode, and in 4.4% of patients admitted with chest pain. The chest pain category of monitored patients was particularly low yield with regard to clinical utility. Electrocardiographic monitoring identified cardiac arrhyth- mia in only a fraction of patients (0.3%) admitted for chest pain. The authors concluded that the overall clinical value of cardiac monitoring is often overestimated by physicians; furthermore, they felt that ECGM was unnecessary in many low-risk patients in this study, particularly the patient with chest pain.

Thus, inpatient electrocardiographic monitoring provides little clinical information which alters patient management during hospitalization, particularly in lower-risk individuals. The low-risk chest pain patient is one group that does not appear to benefit significantly from such monitoring; the other inpatient groups demonstrate a modest benefit at best from ECGM. Clearly, the identification of specific patient groups-both low- and high-risk categories-would be quite helpful in determining the need for electrocardio- graphic monitoring in the inpatient setting.

Risk stratification and electrocardiographic monitoring patient selection

Although it is reasonably well established that electro- cardiographic monitoring is not particularly useful for low- risk patients, an easily applied identification method which could be used in the ED does not yet exist. Over the past several decades, a number of chest pain risk stratification protocols have been independently developed for this

purpose. In 1991, the American College of Cardiology (ACC) published a policy statement regarding recommenda- tions for ECGM [15]. These guidelines broke down patients into 3 different categories based on risk. Patients with a high risk for developing a Life-threatening arrhythmia were categorized as class I and most if not all patients in this group warranted the use of telemetry. Class I included patients (1) suspected of having an acute MI based on clinical evaluation or ECG findings; (2) having an operation; (3) who had recent cardiac surgery or implantation of an automated internal cardioverter defibrillation system; (4) who have recently been resuscitated; (5) with an unstable cardiac rhythm such as Mobitz type II heart block or worse; (6) who are hemodynamically unstable; (7) who are being mechani- cally ventilated; (8) who have been poisoned with chemicals that may produce cardiac arrhythmias; (9) who are receiving loading doses of type I or type III Antiarrhythmic drugs to treat potentially fatal arrhythmias; (10) who just had a percutaneous transluminal coronary angioplasty; (11) with unstable angina; (12) who have high-risk coronary artery lesions; (13) who were recently treated with Catheter ablation for a cardiac arrhythmia.

Patients were allocated to class II if it was felt they may benefit from monitoring and included the following: (1) individuals who had an MI more than 3 days ago; (2) patients who have potentially fatal dysrhythmias days after initially controlling these dysrhythmias; (3) patients who are determined to be at elevated risk for cardiac or respiratory arrest or the sudden onset of hypotension based on physician judgment; (4) persons with clinically significant non-life- threatening arrhythmias (such as atrial fibrillation) that are deemed to be at an enhanced risk for proarrhythmic effects during initial treatment with a type I or III antiarrhythmic drug; (5) patients who have possible or proven hemodyna- mically significant paroxysmal tachyarrhythmias or bradyar- rhythmias; (6) patients experiencing the acute phase of pericarditis when myocarditis is not clinically evident; (7) patients being worked up for unexplained syncope or other transient neurologic signs or symptoms that may be secondary to a cardiac arrhythmia; (8) individuals who just underwent percutaneous transluminal coronary angioplasty;

(9) during the first 48 to 72 hours after patients are implanted with a permanent pacemaker; (10) patients who are stable after cardiac surgery.

Low-risk class III patients could go without monitoring (Table 1) and included the following: (1) postoperative patients that are likely to be low risk; (2) obstetric patients without substantial medical problems; (3) individuals with a terminal illness who are not eligible for treatment of arrhythmias (such as do-not-resuscitate patients); (4) persons who have undergone uncomplicated coronary angiography;

(5) patients with chronic and stable atrial fibrillation; (6) patients experiencing stable asymptomatic premature ven- tricular contractions or nonsustained ventricular tachycardia;

(7) patients with underlying cardiac disease that has been stabilized who have demonstrated no arrhythmias after 3

No ST changes, no arrhythmias, no hemodynamic instability, no history of cardiac disease, age b55 years, no need for general anesthesia within 24 hours of presentation.

No ACS on ECG,

systolic blood pressure N100 mm Hg, no rales auscultated above lung bases, no history of unstable ischemic heart disease.

Negative initial biomarker, initial Goldman score very low to moderate risk. Normal or unchanged initial ECG, normal initial biomarker, no recurrent chest pain requiring therapy, and no life-threatening arrhythmias.

Normal or unchanged initial ECG, normal initial biomarker.

Normal or unchanged initial ECG, normal initial biomarker.

Symptoms ACI-TIPI calculation suggestive of printed onto ECG. acute ischemia.

Selker et al

[25] (1998)

Suspected acute coronary syndrome.

Chest pain.

Kelly and Kerr

[3] (2001)

Kelly and Kerr

[4] (2004)

Kelly and Edmonds Chest pain.

[23] (1997)

Sultana et al

[24] (2002)

Chest pain.

Hollander et al

[21] (2004)

Goldman et al [18] Chest pain. (1996) Durairaj

et al [19] (2001)

Nagy et al

[28] (2001)

Normal ECG findings, normal biomarker.

Low-risk criteria

Type of patient

Atypical chest pain with no associated symptoms.

blunt chest trauma.

Snider et al

[11] (2002)

Table 1 Patients who can go without cardiac monitoring

Study

consecutive days of ECGM. The utility of these guidelines was assessed in a number of studies with conflicting results. One such assessment was undertaken by Estrada et al [16]. The authors conducted a prospective cohort study of 467 patients admitted to a non-ICU cardiac monitoring unit. In this population, 38 (8.1%) monitored patients were transferred to the ICU with electrocardiographic monitoring contributing to this decision in only 5 (13.2% of all ICU transfers) of these individuals. Overall, the majority of patient management was based on clinical observation, with only 5 of 38 ICU patients benefiting from telemetry; importantly, this subset which benefited from electrocardio- graphic monitoring represented only 1.1% of the entire study

group. The researchers note the Financial burden of cardiac monitoring, stating that each monitored bed at their institution costs more than $10000 with their hospital’s 32 monitored beds totaling a cost of more than $320000. Despite this small number of benefited patients, the authors concluded that the use of the 1991 ACA guidelines [15] was of value in the identification of individuals requiring electrocardiographic monitoring.

A later study by Estrada et al [17], however, found that these 1991 ACC guidelines [15] were flawed; they recommended revisions to the guideline. In this prospective cohort study of 2240 patients admitted to non-ICU ECGM units, 61% were assigned to ACC ECGM Guidelines [15] class I, 38% to class II, and 1% to class III. Paradoxically, cardiac dysrhythmias were discovered in only 13.5% of class I patients whereas 40.7% of class II patients and 12% of class III patients had arrhythmias. In addition, ECGM resulted in an alteration of management in 3.4% of class I, 12.7% of class II, and 4% of class III patients. The authors concluded that the guidelines would serve as a more accurate risk stratification tool if they were restructured so that patients with chest pain were moved from class I into class II; furthermore, they recommended that patients presenting with arrhythmias were relocated from class II to class I.

In an effort to create a more accurate set of guidelines, in 2004 Drew et al [9] published an American Heart Association statement describing a risk stratification tool aimed at the identification of patients most likely to benefit from ECGM. The risk stratification tool developed by Drew et al separates patients into 3 different levels, or classes, of concern-not dissimilar from the ACC attempt in 1991 [15]. Class I patients require electrocardiographic monitoring with definite potential for benefit, whereas class II individuals may benefit from ECGM; class III patients are considered low risk and do not need to be monitored. Class I encompasses patients at risk for potentially fatal cardiac arrhythmias who should be monitored at all times. Patients allocated to class I include the following situations and scenarios: (1) resuscitated cardiac arrest; (2) the early phase of ACS; (3) recently diagnosed high-risk coronary lesions;

(4) acute heart failure/pulmonary edema; (5) recent cardiac procedures (eg, cardiac surgery, percutaneous coronary intervention with complication, automatic defibrillator lead, or pacemaker); (6) requirement for procedure (eg, Temporary pacemaker, intra-aortic balloon counterpulsation, and con- scious sedation or anesthesia); and (7) potentially unstable dysrhythmias (AV block, Wolff-Parkinson-White syndrome with rapid rate, and long-QT syndrome with ventricular arrhythmia). Patients in the class II category include individuals with the following features: (1) 24 to 48 hours post-AMI; (2) chest pain with initial nondiagnostic 12-lead ECG and negative cardiac biomarker; (3) syncope; (4) recent Procedure performance (pacemaker implantation, arrhythmia ablative therapy, and uncomplicated coronary angiography);

(5) subacute congestive heart failure; and (6) adjustment of anti-arrhythmic medications. All low-risk patients who do

not fit into the first 2 categories are placed into class III. Thus far, these 2004 Guidelines [9] are merely expert opinion-it must be stressed that this material has yet to be validated by scientific investigation.

Empirically validated risk stratification protocols

Goldman et al [18] have developed the most widely studied risk stratification protocols thus far, examining more than 15000 patients in one study alone. The Goldman criteria risk stratifies patients with chest pain suspected of ACS into high-, moderate-, low-, and Very low risk categories based on 5 variables. The factors associated with an increased risk are 12-lead ECG suggestive of AMI (ST-segment elevation or Q waves), 12-lead ECG suggestive of acute ischemia (ST- segment depression or T-wave inversion), systolic blood pressure less than 100 mm Hg, rales noted bilaterally above the lung bases, and pain worse than prior angina or the same as pain experienced with a prior MI (Fig. 1). The authors defined a major event as one that required intensive care management, which included the following: (1) ventricular fibrillation; (2) cardiac arrest; (3) new Complete heart block;

(4) insertion of a temporary pacemaker; (5) emergency cardioversion; (6) cardiogenic shock; (7) use of an intraaortic balloon pump; (8) intubation; (9) recurrent ischemic chest pain requiring coronary-artery bypass grafting or percuta- neous transluminal coronary angioplasty within 72 hours after admission or cardiac catheterization followed by coronary-artery bypass grafting or percutaneous transluminal coronary angioplasty before discharge.

The Goldman study [18] revealed interesting trends, validating the prediction rule. For example, in the first 12 hours of hospitalization, the rate of new Major adverse events occurring in each of the categories was 0.2% (very low), 0.5% (low), 1.1% (moderate), and 7.6% (high). Risk diminished significantly after the first 24 hours with 41% of major adverse events occurring in the first 12 hours and 62% of major events happening in the first 24 hours among the initial set of patients studied.

Several other studies have validated the Goldman protocol [18]. Durairaj et al [19] performed a prospective cohort study of 1033 consecutive ED patients admitted to a telemetry unit, 677 (65.5%) of whom presented with chest pain. Among all patients studied, major cardiac complication rates were 2% in high-risk patients, 1.5% each in moderate- and low-risk patients, and 0.2% in very low risk patients. An earlier study of 207 patients conducted by Reilly et al [20] also found the prediction rule to be useful for triaging patients. A prospective cohort study by Hollander et al [21] examined the efficacy of a new risk stratification method combining the Goldman protocol [18,22] with the use of cardiac biomarkers. A total of 1750 patients with chest pain were admitted from the ED to non-ICU electrocardiographi-

Fig. 1 Goldman et al [18] risk stratification algorithm.

cally monitored beds. Out of this group, 1029 patients (58.8%) were determined to be low-risk based on the Goldman score [18,22]. The authors concluded that the new protocol was satisfactory as none of the low-risk monitored patients had a significant arrhythmic event.

Kelly and Kerr [3] evaluated the safety of an admission protocol for patients with suspected ACS. According to the protocol, patients were allocated to unmonitored beds if they were pain free after therapy, demonstrated a normal or unchanged 12-lead ECG, and had normal initial biomar- kers in the ED. In a follow-up study, Kelly and Kerr [4] performed a prospective cohort study of patients with chest pain who were admitted to either monitored or unmoni- tored beds, testing their initial protocol [3]. These investigators found that their admission criteria to an unmonitored bed were predictive of a stable hospital course from a rhythm perspective. A similar protocol, conceived by Kelly and Edmonds [23], was examined by Sultana et al [24] in a retrospective study. Patients were defined as low risk and assigned to unmonitored beds if they had normal initial biomarker results and a normal or unchanged 12-lead ECG, no recurrent chest pain requiring intravenous medication, and no life-threatening arrhyth- mias during the ED stay. In these 3 studies, the vast majority of patients stratified to nonmonitored beds did not experience adverse outcomes. Among those who did, none of the suboptimal outcomes could have been prevented by ECGM. The authors of these studies found that low-risk patients with suspected ACS can be safely managed in unmonitored beds.

Assessment of physician triage decisions

Several researchers have compared the admission decisions made by physicians to those of risk stratification protocols. One was a prospective cohort study completed by Selker et al [25] which analyzed 10689 patients presenting to the ED at 10 different hospitals with symptoms suggestive of ACS, with 5951 functioning as controls while 4738 comprised the group. The model used is known as the Acute cardiac ischemia Time-Insensitive Predictive Instrument (ACI-TIPI), which predicted the probability of a patient with chest pain to have ACS. Variables used by the ACI-TIPI included age, sex, and 12- lead ECG findings (Q wave, ST-segment abnormalities, or T-wave inversion). These researchers compared the rates of patient allocation to the ICU, ECGM unit, nonmonitored unit, or home by emergency physicians between the 2 groups in an effort to determine whether the ACI-TIPI was useful in decreasing the need for unnecessary telemetry without increasing the frequency of adverse outcomes. In hospitals with relatively high-capacity telemetry wards, the prediction rule lowered admission to telemetry beds from 68% to 59% with no difference in rates of in-hospital complications, need for rehospitalization, and 30-day mortality between the control and ACI-TIPI groups. In- hospital complications included (1) hypotension; (2) ventricular tachycardia or fibrillation; (3) cardiac arrest;

(4) recent cardiopulmonary resuscitation; (5) recurrent myocardial infarction; (6) administration of Intravenous nitroglycerine; (7) defibrillation; and (8) intraaortic balloon pump. The authors concluded that ACI-TIPI appears to be useful in reducing unnecessary admissions to monitored units, potentially resulting in 200000 fewer unwarranted hospitalizations annually in the United States, without increasing the risk of adverse outcomes.

Several additional studies have been done to assess the triage decisions made by admitting physicians. Gomberg- Maitland et al [26] conducted a retrospective analysis of data collected from the Global Unstable Angina Registry and Treatment Evaluation Registry. Data from 2857 patients were used; patients were retrospectively stratified based on Thrombolysis in Myocardial Infarction risk score into low- (0-2), intermediate- (3-4), and high- (5- 7) risk categories. Patients were given 1 point toward their TIMI score for each of the following: age greater than 65, more than 3 cardiac risk factors, known coronary artery stenosis greater than 50%, use of aspirin in the past 7 days, a recent episode of severe angina, elevated cardiac biomarkers, and ST-segment changes greater than 0.5

mm. When assessing the number of high-, intermediate-, and low-risk patients admitted to monitored and unmoni- tored beds, the authors noted surprising results-essen- tially, physician judgment without the benefit of a risk- stratification protocol resulted in the allocation of more high-risk patients to unmonitored beds than to monitored locations. In addition, more procedures were necessary for

patients on nonmonitored wards than for patients in the telemetry unit or CCU.

Physician judgment regarding patient admission location was also examined by Reilly et al [27]. The authors created 20 written case presentations detailing patients presenting to the ED with chest pain; they compared the admission decisions of 147 physicians (68 attending or fellow physicians and 79 senior resident physicians) to that of the Goldman protocol [18]. A total of 2867 case scenarios were completed. Overall, physicians agreed with the prediction rule in 58% of the cases. Physicians agreed with the prediction rule 83% of the instances when CCU admission was recommended. Yet they only agreed 50% of the time when the prediction protocol suggested admission to the telemetry unit. Interestingly, cardiologists only agreed with the prediction rule 51% of the cases whereas emergency medicine physicians (59%) and internal med- icine physicians (58%) were more likely to be in agreement. Interphysician agreement was relatively poor, ranging from 50% to 59%. Physicians triaged 85% of high-risk patients to monitored beds, whereas the predic- tion protocol accomplished this 96% of the time. Mean- while, physicians were slightly less likely than the prediction rule to allocate low-risk patients to unmonitored beds (38% vs 41%). Overall, physicians were more likely to admit patients to monitored beds than the prediction rule. The complication rate among the patients which the physicians did not admit to a monitored unit, however, was 2.4%, which was 4 times greater than the complication rate of 0.6% determined by the prediction protocol. These studies highlight the need for specific guidelines for ECGM admission by demonstrating the wide variability in physician management decisions and the more efficient use of resources obtained through the utilization of a risk stratification protocol.

Risk stratification of the blunt chest trauma patient

In a different application of electrocardiographic monitor- ing for the admitted patient, Nagy et al [28] reviewed the presentations of patients with blunt trauma requiring inpatient telemetry-basically, the rule-out myocardial contusion process; patients were separated into low- and high-risk categories. High-risk patients were defined as age of 55 years, ST-segment changes, arrhythmias, hemody- namic instability, requirement for general anesthesia within 24 hours of presentation, or a past history of significant cardiac disease; low-risk patients lacked all of these criteria. The study included a total of 315 patients with 144 (45.7%) considered high risk and 171 (54.3%) categorized as low risk. High-risk patients were admitted to the ICU, whereas patients without any of these risk factors were considered low risk and were admitted to an unmonitored ward

Financial impact“>where they received a follow-up 12-lead electrocardiogram

24 hours later. Four patients in the low-risk group demonstrated new abnormalities on the subsequent ECG, but none of these patients had any sequelae from their blunt chest trauma. Meanwhile, 6 of the high-risk patients needed treatment for arrhythmias, pulmonary edema, or hypoten- sion, and one of these patients died. Although high-risk patients warrant the use of inpatient telemetry, the authors concluded that patients who sustain blunt chest trauma who present to the ED with a normal electrocardiogram, normal blood pressure, and no dysrhythmias do not require inpatient cardiac monitoring.

The impact of unnecessary cardiac monitoring

Overcrowding

Although the “pathophysiology” of ED overcrowding is multifactorial, many authorities note a lack of inpatient bed space, especially ICU and ECGM beds, as a major contributor to the problem [29]. In a survey of 575 ED directors, “hospital bed shortage” was identified as the second most significant reason for ED overcrowding [30]. During the course of the study conducted by Durairaj et al [19], an average of 3 patients a day were denied admission to the cardiac monitoring ward owing to a lack of beds whereas an average of 4 of the 12 telemetry beds were occupied by very low risk patients. Naturally, it follows that a reduction in the number of patients who require monitored beds will help to ease the problem of ED overcrowding. Importantly, lengthened ED stay of patients with non-ST-elevation myocardial infarction has been associated with an increased risk of recurrent myocardial infarction and a diminished likelihood of receiving guide- line-recommended therapies [31]. There is no question that ameliorating the problem of ED overcrowding will produce desirable results.

Resource demand

Overburdened telemetry wards are likely to miss more adverse events recorded by telemetry. Peterson et al [32] prospectively compared telemetry reports done at a small community hospital without the use of a dedicated monitor watcher to the reports of an academic cardiology ward equipped with concurrent full monitoring for the patients cared for at the community hospital. The community hospital missed several significant events that the academic cardiol- ogy service noticed, including multiple pauses for longer than 2 seconds, numerous episodes of nonsustained ventricular tachycardia, and several instances of supraven- tricular tachycardia. In a similar study, Atzema et al [33] conducted a prospective cohort study assessing the utility of

continuous electrocardiographic monitoring which involved 72 patients, 56% of whom were categorized as very low risk by Goldman risk criteria [18]. Only 0.2% of alarms resulted in a change in patient management and 99.4% of alarms that sounded were false.

These studies highlight some important points. Physicians often overestimate the value of electrocardiographic mon- itoring; they must realize that such surveillance is not full proof, especially outside of an adequately staffed unit [14,32]. In addition, overuse of telemetry for low-risk patients increases the likelihood that an alarm which is sounding is a false one, a situation that can desensitize monitor watchers and increase the risk of missing a meaningful alarm [1,33]. If low-risk patients who are unlikely to benefit from monitoring can be identified and allocated to unmonitored beds, limited telemetry resources will not be stretched so thin and will be less likely to miss important adverse events.

Financial impact

Inefficient triage of patients with chest pain to monitored beds imposes a financial cost on hospitals and patients. Bayley et al [34] conducted a prospective cohort study of 817 patients with chest pain over the age of 24 years who were admitted to a monitored bed from the ED. The study focused on the extra cost to the hospital of a lengthened ED stay for patients waiting to be admitted to the cardiac telemetry unit. The annual Opportunity cost in lost hospital revenue for patients with chest pain waiting for an ECGM bed was determined to be $168300 or about $204 for every patient who waited longer than 3 hours for a hospital bed. The authors believe their study underestimated the true opportu- nity cost of lengthened ED stays in that the analysis did not consider lost revenue from patients who Left without being seen, Ambulance diversions to other EDs, and other intangible factors. An earlier study by Bayley et al [35] produced similar results.

Krochmal and Riley [36] discussed the financial burden of ED overcrowding for patients, estimating the total extra cost imposed on 26020 patients seen in an ED from 1988 to 1990 to be approximately $6.8 million. The increased cost was mainly due to an average increased length of stay experienced by admitted patients who waited longer than 24 hours in the ED. These studies highlight the financial benefit of allocating patients with low-risk chest pain to unmonitored beds.

Medicolegal implications

The threat of Malpractice litigation likely represents one of the biggest obstacles to the widespread implementation of a standardized chest pain risk stratification scheme and more appropriate, limited application of electrocardiographic monitoring. Although physicians may feel wary about

triaging fewer patients to ECGM units, studies demonstrate that risk stratification protocols not only reduce unnecessary use of monitoring but also decrease the incidence of adverse events [26,27]. In addition, even without the benefit of a risk stratification rule for ECGM identification, Selker et al [25] found that physicians can safely adapt to a limited availability of ICU bed resource without increasing the likelihood of mortality.

Conclusion

Although electrocardiographic monitoring is undoubtedly useful for certain high-risk patients, it is also quite clear that the opposite can be stated regarding the low-risk patient. In fact, simple recognition of the low-risk patient would substantially reduce unnecessary electrocardiographic mon- itoring (Table 1). Certainly, the overuse of cardiac telemetry monitoring has considerable untoward consequences.

References

1. Cusack RJ, Coutts JF. Do not be alarmed, the patient is monitored. Crit

Care 2003;7:349-50.

1. Selker HP, Griffith JL, Dorey FJ, D’Agostino RB. How do physicians adapt when the coronary care unit is full? A prospective multicenter study. JAMA 1987;257:1181-5.
2. Kelly AM, Kerr D. It is safe to manage selected patients with acute coronary syndromes in unmonitored beds? J Emerg Med 2001;21: 227-33.
3. Kelly AM, Kerr D. Clinical features in the emergency department can identify patients with suspected acute coronary syndromes who are safe for care in unmonitored Hospital beds. Intern J Med 2004;34: 594-7.
4. Julian DG, Valantine PA, Miller GG. Disturbances of rate, rhythm and conduction in acute myocardial infarction. Am J Med 1964;37: 915-27.
5. Zoll PM, Linethal AJ, Gibson W, et al. Termination of ventricular fibrillation in man by externally applied countershock. N Engl J Med 1956;254:727-32.
6. Fineberg HV, Scadden D, Goldman L. Care of patients with a low probability of acute myocardial infarction. cost effectiveness of alternatives to coronary-care-unit admission. N Engl J Med 1984;310: 1301-7.
7. Goldman L. Coronary care units. A perspective on their epidemiologic impact. Int J Cardiol 1982;2:284-7.
8. Drew BJ, Califf RM, Funk M, et al. Practice standards for electrocardiographic monitoring in hospital settings: an American Heart Association scientific statement from the councils on cardio- vascular nursing, clinical cardiology, and cardiovascular disease in the young. Circulation 2004;110:2721-46.
9. Hollander JE, Valentine SM, McCuskey CF, et al. Are monitored telemetry beds necessary for patients with nontraumatic chest pain and normal or nonspecific electrocardiograms? Am J Cardiol 1997;79:1110-1.
10. Snider A, Papaleo M, Beldner S, et al. Is telemetry monitoring necessary in low-risk suspected acute chest pain syndromes? Chest 2002;122:517-23.
11. Saleem MA, McClung JA, Aronow WS, Kannam H. Inpatient telemetry does not need to be used in the management of older

patients hospitalized with chest pain at low risk for in-hospital coronary events and mortality. J Gerontol 2005;60A:605-6.

1. Schull MJ, Redelmeier DA. Continuous electrocardiographic monitor- ing and Cardiac arrest outcomes in 8,932 telemetry ward patients. Acad Emerg Med 2000;7:647-52.
2. Estrada CA, Rosman HS, Prasad NK, et al. Role of telemetry monitoring in the non-intensive care unit. Am J Cardiol 1995;76: 960-5.
3. Emergency Cardiac Care Committee Members. Recommended guide- lines for in-hospital cardiac monitoring of adults for detection of arrhythmia. J Am Coll Cardiol 1991;18:1431-3.
4. Estrada CA, Prasad NK, Rosman HS, et al. Outcomes of patients hospitalized to a telemetry unit. Am J Cardiol 1994;74:357-62.
5. Estrada CA, Rosman HS, Prasad NK, et al. Evaluation of guidelines for the use of telemetry in the non-intensive-care setting. J Gen Intern Med 2000;15:51-5.
6. Goldman L, Cook EF, Johnson PA, et al. Prediction of the need for intensive care in patients who come to emergency departments with acute chest pain. N Engl J Med 1996;334:1498-504.
7. Durairaj L, Reilly B, Das K, et al. Emergency department admissions to inpatient cardiac telemetry beds: a prospective cohort study of risk stratification and outcomes. Am J Med 2001;110:7-11.
8. Reilly BM, Durairaj L, Husain S, et al. Performance and potential impact of a chest pain prediction rule in a large public hospital. Am J Med 1999;106:285-91.
9. Hollander JE, Sites FD, Pollack Jr CV, et al. Lack of utility of telemetry monitoring for identification of cardiac death and life-threatening ventricular dysrhythmias in low-risk patients with chest pain. Ann Emerg Med 2004;43:71-6.
10. Goldman L, Cook EF, Brand DA, et al. A computer protocol to predict myocardial infarction in emergency department patients with chest pain. N Engl J Med 1988;318:797-803.
11. Kelly AM, Edmonds E. Managing potentially ischaemic chest pain and coronary care beds effectively. Aust Health Rev 1997;29:40-8.
12. Sultana RV, Kerr D, Kelly AM, et al. Validation of a tool to safely triage selected patients with chest pain to unmonitored beds. Emerg Med (Fremantle) 2002;14:393-9.
13. Selker HP, Beshansky JR, Griffith JL, et al. Use of the acute cardiac ischemia time-insensitive predictive instrument (ACI-TIPI) to assist with triage of patients with chest pain or other symptoms suggestive of acute cardiac ischemia. A multicenter, controlled clinical trial. Ann Intern Med 1998;129:845-55.
14. Gomberg-Maitland M, Murphy SA, Moliterno DJ, Cannon CP. Are we appropriately triaging patients with unstable angina? Am Heart J 2005; 149:613-8.
15. Reilly BM, Evans AT, Schaider JJ, Wang Y. Triage of patients with chest pain in the emergency department: a comparative study of physicians’ decisions. Am J Med 2002;112:95-103.
16. Nagy KK, Krosner SM, Roberts RR, et al. Determining which patients require evaluation for blunt cardiac injury following blunt chest trauma. World J Surg 2001;25:108-11.
17. Derlet RW, Richards JR. Overcrowding in the nation’s emergency departments: complex causes and disturbing effects. Ann Emerg Med 2000;35:63-8.
18. Derlet RW, Richards JR, Kravitz RL. Frequent overcrowding in US emergency departments. Acad Emerg Med 2001;8:151-5.
19. Diercks DB, Roe MT, Chen AY, et al. Prolonged emergency department stays of non-ST-segment-elevation myocardial infarction patients are associated with worse adherence to the American College of Cardiology/American Heart Association guidelines for management and increased adverse events. Ann Emerg Med 2007; 50:489-96.
20. Peterson MC, Whetten DK, Renlund DG, Coletti A. Sensitivity of rhythm disturbance detection by community hospital telemetry. Ann Noninvasive Electrocardiol 2002;7:219-21.
21. Atzema C, Schull MJ, Borgundvaag B, et al. Alarmed: adverse events in low-risk patients with chest pain receiving continuous

electrocardiographic monitoring in the emergency department. A pilot study. Am J Emerg Med 2006;24:62-7.

1. Bayley MD, Schwartz JS, Shofer FS, et al. The financial burden of emergency department congestion and Hospital crowding for chest pain patients awaiting admission. Ann Emerg Med 2005;45:110-7.
2. Bayley MD, Schwartz JS, Shofer FS, et al. The financial burden of ED congestion and Hospital overcrowding for chest pain patients awaiting admission. Acad Emerg Med 2002;9:367.
3. Krochmal P, Riley TA. Increased health care costs associated with ED overcrowding. Am J Emerg Med 1994;12:265-6.