|Home | About | Journals | Submit | Contact Us | Français|
To examine the appropriate use of arrhythmia, ischemia, and QTc interval monitoring in the acute care setting.
We analyzed baseline data of the PULSE Trial, a multi-site randomized clinical trial evaluating the effect of implementing ECG monitoring practice standards. Research nurses reviewed medical records for indications for monitoring and observed if arrhythmia, ischemia, and QT interval monitoring were being done on 1,816 patients in 17 hospitals.
Almost all (99%) patients with an indication for arrhythmia monitoring were being monitored, but 85% of patients with no indication were monitored. Of patients with an indication for ischemia monitoring, 35% were being monitored, but 26% with no indication were being monitored for ST-segment changes. Only 21% of patients with an indication for QT interval monitoring had a QTc documented, but 18% of patients with no indication had a QTc documented.
Our data show evidence of inappropriate monitoring: under-monitoring for ischemia and QTc prolongation and over-monitoring for all 3 types of monitoring, especially arrhythmia monitoring.
Although electrocardiographic (ECG) monitoring is the cornerstone of care in hospital cardiac units, few studies1–5 have evaluated how it is used. The American Heart Association (AHA) Practice Standards for ECG Monitoring in Hospital Settings6 specify indications and time frames for ECG monitoring. The purpose of this analysis of baseline data from the Practical Use of the Latest Standards for Electrocardiography (PULSE) Trial is to examine the appropriate use of arrhythmia, ST-segment (ischemia), and QT interval monitoring in current clinical practice.
The PULSE Trial is a 5-year (2008–2013) multi-site randomized clinical trial to evaluate the implementation of the AHA Practice Standards for ECG Monitoring6 on nurses’ knowledge, quality of care including the appropriateness of monitoring, and patient outcomes. The intervention consists of an online ECG monitoring education program and strategies to implement and sustain change in practice, led by nurse champions on each unit. The study takes place in 17 hospitals: 15 in the United States, 1 in Ottawa, Canada, and 1 in Hong Kong, China (Table 1). All hospitals received institutional review board approval. Sites include both academic medical centers and community hospitals. Hospital units involved in the study are primarily for the treatment of cardiac surgical and medical patients. They include both intensive care units (ICUs) with “hard-wire” bedside cardiac monitoring and step-down units with “wireless” telemetry monitoring.
For the baseline quality of care data, our sample consisted of 2,744 observations on 1,816 patients on these adult cardiac units. One of three research nurses, who were experienced ICU nurses with expertise in ECG monitoring, visited each site for 5 days. If time permitted, they observed patients more than once during their 5-day visit. They collected data on the use and appropriateness of monitoring by reviewing the current medical records to determine if the patient had a Class I or II indication for arrhythmia, ischemia, or QT interval monitoring. The AHA Practice Standards6 used the following rating system for indications for ECG monitoring: Class I – indicated in most, if not all, patients; Class II – may be of benefit in some patients, but is not considered essential for all patients; and Class III – not indicated because a patient’s risk of a serious event is so low that monitoring has no therapeutic benefit. The research nurses also observed if arrhythmia, ischemia, and QT interval monitoring were being done. Specifically, they determined whether nurses activated the ST-segment monitoring feature on the monitors or documented the QTc interval in the medical record. The classifications for appropriateness of monitoring are outlined in Table 2. If patients had a Class I indication for a particular type of monitoring, they should be monitored. If patients did not have either a Class I or II indication for a particular type of monitoring, they should not be monitored.
Tables 3–5 outline the Class I indications for arrhythmia, ischemia, and QT interval monitoring and recommendations for how long monitoring should be maintained.7 Patients who have top priority for arrhythmia monitoring are shown in Table 3. The research nurse determined if a patient had a Class I or II indication for arrhythmia monitoring and then observed if the patient was receiving continuous ECG monitoring when she was on the unit.
Patients who have top priority for ST-segment monitoring include those at significant risk of myocardial ischemia which, if sustained, may result in acute myocardial infarction (MI) or extension of the MI (Table 4). It is not appropriate for all patients to be monitored for myocardial ischemia. For example, patients with intermittent ventricular pacing that results in secondary ST/T wave abnormalities and triggers frequent false ST-segment alarms are inappropriate for ST monitoring. In addition, patients with left bundle branch block and those with intermittent right bundle branch block should not be continuously monitored for ST-segment changes.6 Activation of ST monitoring adds another layer of potential false alarms. The research nurse determined if a patient had a Class I or II indication for ischemia monitoring and observed if the ST-segment monitoring feature was activated.
Patients who have top priority for QT interval monitoring are shown in Table 5. The research nurse determined if a patient had a Class I or Class II indication for QT interval monitoring, specifically, whether the patient was started on a QT-prolonging drug, had a primary admission diagnosis of drug overdose, had a heart rate slower than 40 beats per minute or a pause > 2 seconds, or had existing drug-induced long-QT syndrome. She then determined if the nurse had documented the QTc value in the patient’s medical record within the last 24 hours.
The sample of 1,816 patients had a mean age of 65, were 57.4% male, and 79.5% white (Table 6). The most common primary diagnosis was a non-cardiac diagnosis (28.2%), followed by acute MI or rule out MI (16.4%) and arrhythmia or syncope (15.7%). Cardiac procedures were relatively uncommon: 14.8% had cardiac surgery and 10.4% had a percutaneous coronary intervention (PCI). Almost all patients (94.1%) were receiving continuous ECG monitoring.
Table 7 shows the results related to the appropriateness of monitoring. Almost all (98.9%) patients with an indication for arrhythmia monitoring were being monitored. However, 84.9% of patients without an indication for arrhythmia monitoring were also being monitored. Of the patients with an indication for ischemia monitoring, 34.5% were being continuously monitored for ST-segment changes, but 26.3% of patients without an indication for ischemia monitoring were also being monitored for ST-segment changes. Only 21.4% of the patients with an indication for monitoring for QT prolongation had a QTc value documented by nurses in the previous 24 hours. However, 18.1% of patients with no indication for QT monitoring also had a QTc value documented.
Our data show evidence of inappropriate monitoring – both under-monitoring for ischemia and QT interval, as well as over-monitoring for all three types of monitoring. There are several possible reasons for evidence of such widespread inappropriate monitoring.
The observation that there is over-monitoring for arrhythmias could be related to leaving patients on the monitor longer than necessary. Physicians may order monitoring because the patient – who may have no risk for arrhythmias – will be watched more closely due primarily to the increased nurse-to-patient ratios on units with ECG monitoring. In addition, once a patient is admitted to an ICU bed, ECG monitoring is usually automatically initiated. It could be deemed inappropriate to have a patient in the ICU without the bedside ECG monitor on, regardless of the admitting diagnosis.
Others have also reported unnecessary monitoring,1,3,4 and guidelines have been implemented to attempt to focus monitoring on patients most likely to benefit. 2,5,8 These studies have taken place in single hospitals using guidelines from 19912,5 or hospital-specific protocols.8 The PULSE Trial is being conducted at 17 hospitals using more recent standards6 that also incorporate recommendations for ischemia and QT interval monitoring.
Langer et al.9 showed that ST-segment elevation as detected by continuous ST-segment monitoring is an independent predictor of mortality, even after controlling for multiple clinical factors. Our previous research 10,11 revealed that nurses do not monitor for ST-segment changes indicative of ischemia because of lack of physician support for it, numerous false alarms that can be annoying, lack of knowledge about how to use the technology and what to do in response to ST alarms, and because they perceive it to be difficult to use. There was also evidence of over-monitoring for ischemia. There may have been initiatives on units to monitor for ST-segment changes, but nurses may not have taken the time to consider whether a particular patient was truly at risk for ischemia and just activated the ST-segment monitoring feature for all patients. In addition, on some patient care units the ST-segment option may be defaulted to be operational at all times. The nurse would need to turn it off if it is not appropriate for a particular patient, and may not have bothered to turn this feature off.
It is well-known that a prolonged QT interval is an important precursor to the lethal arrhythmia torsades de pointes.12 There was evidence in our study of both under- and over-monitoring for QT-interval prolongation. Possible reasons for not monitoring for QT prolongation include that the physician does not order it or a lack of knowledge on the part of nurses regarding factors placing patients at risk for torsades de pointes or how to use the formula to correct for heart rate. In addition, nurses may perceive that using a correction formula may be too difficult and time-consuming. These nurses may not know about or be resistant to using electronic calipers, if it is option on the ECG monitor. We also observed over-monitoring for QT-interval prolongation. A nurse at one of the sites observed that it was easier to obtain QTc measurements on all patients rather than thinking about whether each individual patient had an indication.
The underutilization of ischemia and QT interval monitoring may be a reflection of a lack of knowledge on the part of nurses. Since its inception in the first coronary care units almost 50 years ago,13,14 continuous ECG monitoring has been primarily focused on the detection of arrhythmias, and arrhythmia interpretation has been the focus of ECG monitoring courses for nurses. Evaluation of ST-segment deviation and measurement of QT interval duration has traditionally been done only as part of an assessment of a 12-lead ECG. It is relatively recently that standards for continuous monitoring of ischemia15 and QT interval prolongation12 have been published. Perhaps with the addition of content on ischemia and QT interval monitoring into academic curriculum and continuing education for nurses, the appropriate use of these types of monitoring will improve.
Some might argue that over-monitoring is preferable to under-monitoring and that we should monitor all patients on cardiac units for arrhythmias, ST-segment changes, and QT interval prolongation. In doing so, we may uncover clinically-relevant abnormalities in patients not considered to be at risk, while not having to search for information needed to decide if a particular patient should have a particular type of monitoring. Conversely, monitoring for ST segment deviations potentially requires additional nursing time in activating this option and responding to false alarms. In the absence of newly available continuous QT interval monitoring, nurses might consider measuring the QT interval and calculating the QTc to take too much time. Lastly, some patients on cardiac units do not have a primary cardiac diagnosis (28.2% of patients in our sample had a non-cardiac primary diagnosis) and may not be appropriate for all three types of monitoring. Studies are needed to determine if there are outcome differences in patients who are over-monitored. Currently, the AHA Practice Standards for ECG Monitoring6 specify Class I, II, and III indications for each type of monitoring. Until research demonstrates otherwise and updated Practice Standards are published, physicians and nurses should be guided by these Practice Standards to decide which patients should have which types of monitoring.
This research has both limitations and strengths. The research nurse was at each hospital for only 5 days, so observed only a snapshot of ECG monitoring practice. There is also the possibility that nurses paid more attention to the quality of their practice because they knew they were being observed (Hawthorne effect). Practice could be worse than we observed. Because of time constraints, the research nurse was often not able to return to check if patients had an indication for monitoring on subsequent days. We defined doing QT interval monitoring as the nurse documenting the QTc in the patient’s medical record. Nurses may not have had a chance to record this value before the research nurse left the unit for the day.
Strengths of this study include the relatively large sample of patients, the inclusion of all cardiac units in 17 hospitals across the United States and internationally, and the prospective design. Expert nurses observed clinical practice in every patient on the unit over consecutive days.
Our data show evidence of inappropriate monitoring: underutilization of ischemia monitoring and failure to monitor for QTc prolongation when indicated, as well as over-monitoring for all 3 types of monitoring, especially arrhythmia monitoring. The next phase of the PULSE Trial is testing whether an online ECG monitoring education program based on the AHA Practice Standards6 and strategies to implement and sustain change in practice will enhance the appropriateness of ECG monitoring.
Funding: National Heart, Lung and Blood Institute (R01 HL81642)
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Marjorie Funk, Yale University School of Nursing.
Catherine G. Winkler, Yale University School of Nursing.
Jeanine L. May, Yale University School of Nursing.
Kimberly Stephens, University of California, San Francisco School of Nursing.
Kristopher P. Fennie, Yale University School of Nursing.
Leonie L. Rose, Yale University School of Nursing.
Yasemin E. Turkman, Yale University School of Nursing.
Barbara J. Drew, University of California, San Francisco School of Nursing.