Anesth Analg 2003;97:1483-1487
© 2003 International Anesthesia Research Society
CRITICAL CARE AND TRAUMA
The Use and Effectiveness of Electrocardiographic Telemetry Monitoring in a Community Hospital General Care Setting
J. Paul Curry, MD*,
C. William Hanson, III, MD FCCM
,
Michael W. Russell, MD,
Cheryl Hanna, RN*,
Gayle Devine, BSN CCRN, and
E. Andrew Ochroch, MD
*Department of Anesthesiology and Critical Care, Hoag Memorial Presbyterian Hospital Newport Beach, CA, and the
Department of Anesthesiology and Critical Care, Hospital of the University of Pennsylvania, Philadelphia, PA
Address correspondence to E. Andrew Ochroch, MD, Department of Anesthesiology, UPHS, 416c Ravdin Ct., 3400 Spruce St., Philadelphia, PA 19104. Address email to ochrocha{at}uphs.upenn.edu
 |
Abstract
|
|---|
The purpose of this study was to determine if rates of telemetry events differ between patients whose monitoring is appropriately "indicated" versus "not indicated" by systematically applying rigorous criteria for appropriateness of electrocardiogram (ECG) telemetry usage. We performed a retrospective cohort study on 1097 telemetry admissions between January 1, 2000 and March 31, 2000. A convenience sample of 218 patients generated 236 telemetry admissions. One-hundred-sixty-two arrhythmic events were detected during 400 "indicated" telemetry days. Nine arrhythmic events were detected during 345 "not indicated" telemetry days. The relative rate for arrhythmic events was significantly different, at P < 0.0001, with the incidence rate ratio of 15 indicating a very large effect size. Consequently, current use of ECG telemetry may not be optimal, and a prospective analysis of the application of rigorous indications for ECG telemetry needs to be undertaken.
IMPLICATIONS: The application of standard criteria to electrocardiogram telemetry admissions found that the majority of abnormal heart rhythms were found when patients met appropriate criteria.
|
Introduction
|
|---|
Several studies have emphasized the risk of serious adverse events faced by the hospitalized population at large (1–6). It is widely perceived that reduction in nurse staffing levels, shifts from clinical to clerical nursing responsibilities, and increases in the acuity of hospitalized patients will make such events intrinsically more likely. The desire to reduce the occurrence of these events has led clinicians to an increasing reliance on technologic aids to patient care outside of the intensive care unit (ICU).
Monitoring of the electrocardiogram by remote telemetry (ECGt) to a central monitoring location is a long-established practice in selected general care areas of most hospitals. Originally developed in the 1960s as a method of reducing dysrhythmic deaths after acute myocardial infarction (MI), ECGt is now applied outside of the ICU to a wide variety of medical and surgical patients in an attempt to reduce the risk of adverse or catastrophic events. In some patients, notably cardiothoracic (CT) surgery patients, the risk of significant dysrhythmic events may be increased enough to warrant ECGt. Until recently, ECGt was the only available option to monitor general care patients by telemetry for prevention of catastrophic events. Requests for ECGt on hospital admission or after transfer from an ICU or a postanesthesia care unit (PACU) have increased beyond capacity in many hospitals. Although criteria have been proposed to define the population that might be expected to benefit from ECGt (7), they are not often applied systematically. There have been no studies addressing the effectiveness of ECGt as it is typically applied to hospitalized patients in general (i.e., those without current ischemia/MI or recent CT surgery).
It was the intent of this study to address two related questions. First, how is ECGt actually used in a community hospital non-ICU setting? Second, will the application of a specified set of criteria for ECGt capture dysrhythmic events of significance without missing clinically important events in patients who did not meet proposed ECGt criteria?
|
Methods
|
|---|
The study was approved by the IRB of the Hoag Memorial Presbyterian Hospital in Newport Beach, CA. The population for study consisted of all patients admitted between January 1, 2000 and March 31, 2000 to a community hospital general care setting where ECGt monitoring was requested and applied. All patients were monitored with a Hewlett Packard Viridia ECG system. A dedicated ECG analyst was monitoring the system, and paper recordings of all arrhythmias were placed into patients’ charts as per hospital protocol. Sinus bradycardia was defined as a sinus node initiated heart rate <50 bpm. Sinus tachycardia was defined as a sinus node initiated heart rate >110 bpm. Supraventricular tachycardia was defined as a nonsinus node initiated heart rate >110 bpm. Ventricular tachycardia (VT) was a ventricular rate >110 bpm with at least 3 beats recorded. Sustained VT was >30 beats of VT. Hemodynamic compromise from an arrhythmia was determined by documented blood pressure changes that led to interventions or changes in patient status (e.g., ischemia, decreased mental status, transfer to ICU).
Because there were no data available on the rates of arrhythmic events for patients who met or did not meet established ECGt criteria, no formal power analysis was attempted. Instead, a large convenience cohort sample was collected in the following manner: of 1097 patient admissions with ECGt monitoring, 224 were chosen for analysis by starting at the first chart and choosing every 30th chart. Six of these patients had "do not resuscitate" orders in effect at the time of telemetry monitoring and were excluded from analysis. This left 218 patients for full analysis.
A given patient may have had more than one ECGt application during the period of study. Thus, these 218 patients (156 medical and 62 surgical) accounted for 236 telemetry admissions. The unit of analysis was a day of ECGt monitoring. Criteria derived from the American College of Cardiology (ACC) Policy Statement on Recommended Guidelines for In-Hospital Cardiac Monitoring (7) with a modification by a Hoag medical staff consensus panel consisting of cardiologists, intensivists, internists, and anesthesiologists were used to define "indicated" days of ECGt. There was only one change made: the panel reduced to 2 days (48 h) the 1991 ACC guideline of 3 days as an appropriate period of ECGt application in patients admitted with primary cardiac disease because of the negligible rate of dysrhythmias detected after 48 h of ECGt that was found during a recent internal quality improvement process.
The modified ACC criteria for appropriate application of ECGt (i.e., "indicated" and "not indicated") are listed in Table 1. On days when these criteria were not met ECGt was considered not indicated. After 48 h of monitoring without ECGt events, days were considered not indicated. Therefore, patients could have ECGt admissions that were completely indicated, completely not indicated, or mixed. The mixed category was developed for those patients whose initial placement on ECGt met our criteria, but who no longer met criteria for monitoring after they stayed dysrhythmia free for 48 h.
Data were obtained from a retrospective review of ECGt recordings ("strips") and medical record entries. ECG events were prospectively defined as described above, and are shown along with the number of such events identified in Table 2.
Demographic data were analyzed using one-way analysis of variance, the Mann-Whitney U-test, and Fisher’s exact test (two tailed) for continuous, ordinal, and proportional data, respectively. The occurrence of significant ECG events was grouped by presence or absence of the modified ACC indications and compared on a per day of ECGt basis. The potential difference in telemetry event rates was examined using the incidence rate ratio. A logistic regression analysis was performed to explore the potential role of confounding and interaction among the variables. All data were entered into a Microsoft Excel database (Microsoft; Redmond, WA) and all statistical analysis was done using STATA 7 (STATA CORP, College Station, TX).
|
Results
|
|---|
The demographic analysis is shown in Table 3. Table 4 lists the reasons for ECGt listed by attending physicians. There are no statistically significant differences in the proportions of the characteristics of age, male sex, height, or weight among the patients whose ECGt days were indicated, not indicated, or mixed. ASA physical status on admission to the hospital is shown as a marker of the in-hospital risk of morbidity and mortality of the patient (8,9). There are no statistically significant differences in the ASA physical status among the patients whose ECGt days were indicated, not indicated, or mixed. Consequently, there do not seem to be any obvious biases in baseline patient characteristics originating from our scheme to choose patients retrospectively.
The 236 telemetry admissions generated 745 telemetry days for analysis. The breakdown of ECGt into indicated and not indicated days is presented in Table 5. For analysis, the mixed days and events were divided into indicated or not indicated based on whether they met the criteria when their event (if any) was recorded.
Of the ECGt admissions analyzed, slightly less than half met the agreed-on indications at no time (29%) or during any day (17%) of the ECGt application. There were 171 dysrhythmic events detected over the 745 total days of telemetry monitoring (Tables 2 and 5 ). For patients who met our telemetry criteria (indicated), 162 telemetry events were recorded during 400 telemetry days. For patients who did not meet our telemetry criteria (not indicated), 9 telemetry events were recorded during 345 telemetry days. As shown in (Table 5), 95% of events occurred on ECGt indicated days; only 5% occurred in the absence of reasonable indications. All of these 9 events were comparatively minor and self-limited, and none required specific therapy. The rate of events occurring on indicated versus not indicated days is highly significantly different, with a P < 0.0001. The effect size is large, with an incidence rate ratio of 15.53 (95% confidence interval, 7.98–34.56). The logistic regression model did not indicate any confounding or effect modification attributable to sex, age, ASA physical status, medical versus surgical service, weight, or height. Consequently, the capture of telemetry events is extremely highly correlated with appropriate application of the criteria listed in Table 1.
|
Discussion
|
|---|
From this retrospective chart review, it is apparent that ECGt is frequently applied in the absence of any appropriate indications. Nearly a third (29%) of ECGt applications in our study were without indication at any time during the monitoring period. Almost one-fifth (17%) of ECGt applications continued beyond the 48-hour period where they would have clinical utility as per our internal quality improvement process. Thus, nearly 50% of all ECGt patient-days are of doubtful clinical utility. The economic and personnel costs associated with overuse of ECGt are likely to be significant. In addition, it may well be that 50% of all patients on ECGt at any given time are not benefiting from its application. The impact of reduced ECGt availability on appropriate in-hospital placement for patients who might have reasonable short-term indications for it (e.g., postoperative CT surgical patients, emergency department patients awaiting admission for suspected myocardial ischemia/MI, or patients with newly diagnosed primary dysrhythmias who are otherwise stable) may be substantial. Finally, the ability to move otherwise stable patients from the ICU, operating room (OR), or PACU to a general care floor may be significantly impaired. In the absence of ECGt availability, an admitting physician may feel that he or she has no recourse but to leave the patient in the ICU when only the ECG needs to be monitored and no other active intervention is required. Conversely, intensivists are often confronted with the argument that ECGt is required for discharge of a patient from the ICU with no other justification than the fact that the patient is somehow at risk, despite the lack of any reasonable indication for such monitoring. The net result is a trickle-down effect that potentially delays appropriate patient flow into and out of the emergency room, OR, PACU and ICU.
There are significant limitations to this study. First, it is a retrospective chart review of a comparatively small number of patients. This nonrandomized retrospective review may over- or underestimate the true effect of the application of appropriate standards to ECGt. Furthermore, even though there was a huge disparity in the number of arrhythmias discovered between the groups, no outcomes or changes in management were examined to allow us to determine the impact of the arrhythmias on patient care and outcome. Although this study is useful in describing existing utilization patterns at a single hospital, the assumption that the application of a more rigorous, criteria-driven protocol for admission to ECGt beds will be safe and also improve the efficiency of care requires prospective confirmation. This study does, however, provide the necessary preliminary data to make such an undertaking reasonable.
This study did not address the potential role for detecting myocardial ischemia during telemetry monitoring. Although any ECG signal that is continuously monitored has the potential to monitor ischemia, there are several difficulties encountered in this practice. The fidelity of the signal transmissions needs to be of a high quality, and the frequency response of the system needs to be appropriate for ST segment monitoring (10). The telemetry system should be able to monitor at least leads II and V5, as this was shown to potentially capture up to 97% of all ECG-detectable ischemic changes during exercise testing (11). Finally, the greatest difficulty in telemetry-based ischemia monitoring probably comes from the continual position change of the patients, which results in a change in ST segment baseline and makes continuous monitoring difficult.
The population in our study is quite heterogeneous, representing a cross-section of medical/surgical patients in a general care hospital. We used the ASA physical status as a measure of patient severity to overcome the potential confounding inherent in the diversity of our patient population. The ASA physical status was not developed for this purpose; however, there is evidence that most global indicators of patient sickness or severity can be used in retrospective cohort studies to minimize confounding (12). Consequently, whether our results are generalizable to other hospitals and patient populations remains to be seen. This study does not address the considerable effort that might be required to implement a system of screening for ECGt indications, both at the time of the original request for ECGt and daily during a particular ECGt monitoring period. Despite these limitations, our data strongly suggest that the application of ECGt may not be appropriate in a significant proportion of cases.
|
Conclusions
|
|---|
Ours is the first study to quantitatively demonstrate the degree of potential over-utilization of ECGt, a technology made more seductive by its apparent simplicity and its logical but untested ability to aid intervention during rare but significant occurrences. The facts that many patients, particularly postoperative patients, may be at greater risk of primary respiratory compromise (13,14) and that ECGt events may in those circumstances represent secondary rather than primary events (15) make the choice of ECGt as the appropriate monitoring modality questionable. Monitoring systems based on other vital signs—notably, continuous pulse oximetry—may provide a more appropriate monitoring strategy for those patients. This hypothesis is currently being investigated.
ECGt seems to be widely over-applied in community hospital general (ward) care settings. The retrospective evidence presented here suggests that, in the absence of reasonable criteria for its use, ECGt is extremely unlikely to detect arrhythmic events that warrant diagnostic or therapeutic intervention. Suggested criteria exist for appropriate, indicated use of ECGt (7). Logically, use of such criteria to justify the initial application and subsequent continuation of ECGt should reduce inappropriate use while improving access for those patients who might stand to benefit. Further, the reduction in requests for ECGt where no indication reasonably exists should safely allow a more appropriate location and movement of patients within the hospital and improve the efficiency of care in other areas. These hypotheses are deserving of prospective evaluation.
|
Acknowledgments
|
|---|
Supported by an unrestricted grant from Tyco International, Ltd.
|
References
|
|---|
- Kohn LT, Corrigan JM, Donaldson MS, eds. To err is human: building a safer health system. Washington, DC: National Academy Press, 1999.
- Brennan TA. The Institute of Medicine report on medical errors: could it do harm? N Engl J Med 2000; 342: 1123–5.[Free Full Text]
- IOM Quality of Health Care in America Committee. The Institute of Medicine report on medical errors: misunderstanding can do harm. MedGenMed 2000; 2: 1–5.
- McDonald CJ, Weiner MW, Hui SL. Deaths due to medical errors are exaggerated in the Institute of Medicine report. JAMA 2000; 284: 93–5.[Free Full Text]
- Leape LL. Institute of Medicine medical error figures are not exaggerated. JAMA 2000; 284: 95–7.[Free Full Text]
- Hayward RA, Hofer TP. Estimating hospital deaths due to medical errors: preventability is in the eye of the reviewer. JAMA 2001; 286: 415–20.[Abstract/Free Full Text]
- ACC policy statement: recommended guidelines for in-hospital cardiac monitoring of adults for detection of arrhythmia. J Am Coll Cardiol 1991; 18: 1431–3.[ISI][Medline]
- Vacanti CJ, vanHouten RJ, Hill RC. A statistical analysis of the relationship of physical status to postoperative mortality in 68,388 cases. Anesth Analg 1970; 49: 564–6.[Free Full Text]
- Cohen MM, Duncan PG. Physical status score and trends in anesthetic complications. J Clin Epidemiol 1988; 41: 83–90.[ISI][Medline]
- Bragg-Remschel DA, Anderson CM, Winkle RA. Frequency response characteristics of ambulatory ECG monitoring systems and their implications for ST segment analysis. Am Heart J 1982; 103: 20–31.[ISI][Medline]
- Blackburn H, Katigbak R. What electrocardiographic leads to take after exercise. Am Heart J 1964; 67: 184–9.[ISI][Medline]
- Schneeweiss S, Maclure M. Use of comorbidity scores for control of confounding in studies using administrative databases. Int J Epidemiol 2000; 29: 891–8.[Abstract/Free Full Text]
- Rosenberg AL, Hofer TP, Hayward RA, et al. Who bounces back? Physiologic and other predictors of intensive care unit readmission. Crit Care Med 2001; 29: 511–8.[ISI][Medline]
- Durbin CG, Kopel RF. A case-controlled study of patients readmitted to the intensive care unit. Crit Care Med 1993; 21: 1547–53.[ISI][Medline]
- Rosenberg J, Dirkes WE, Kehlet H. Episodic arterial oxygen desaturation and heart rate variations following major abdominal surgery. Br J Anaesth 1989; 63: 651–4.[Abstract/Free Full Text]
Accepted for publication May 29, 2003.
This article has been cited by other articles:
|
|
|
|
 
T. Reilly and D. Humbrecht
Fostering Synergy: A Nurse-Managed Remote Telemetry Model
Crit. Care Nurse,
June 1, 2007;
27(3):
22 - 33.
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
