July 2011: Volume 37, Number 7
Low-Risk Chest Pain Patients
Can a Two-Hour Assessment Protocol Send Them Home Safely?
By Genna Rollins
Emergency department overcrowding has confounded hospital administrators and clinicians for years, with no simple solution in sight. In fact, emergency visits increased at twice the rate of growth of the U.S. population between 1997 and 2007, and a recent Government Accountability Office report found that it took more than double the recommended time to see patients with immediate care needs. Given these strains on the healthcare system, the idea of being able to quickly assess and safely discharge even a modest segment of the emergency population is quite appealing.
That’s why a provocative protocol put forth by a consortium of Pacific Rim researchers has generated discussion internationally. The protocol, which involves a 2-hour accelerated diagnostic protocol using a point-of-care (POC) biomarker panel, a risk score, and electrocardiography (ECG) to identify and safely discharge patients with symptoms suggestive of acute coronary syndrome (ACS), one day could lead to much shorter assessment times for a substantial and resource-intensive segment of the emergency population, low-risk chest pain patients (Lancet 2011;377:1077–84). This approach has piqued interest in both the cardiology and emergency medicine fields because each year, patients presenting with symptoms suggestive of ACS account for as many as 10% of hospital emergency cases and up to one-quarter of admissions. Most undergo hours-long assessments, although 75–85% turn out not to have ACS, according to lead investigator Martin Than, MBBS.
“We have to take people who present with chest pain very seriously, because the consequences of getting it wrong are significant. Concern about the possibility of sending someone home to harm is a key driver of clinician behavior, which is essentially to investigate chest pain patients to get the risk as low as possible,” Than explained. “Given that only 20 to 25 percent have acute myocardial infarction, that’s a lot of people being investigated. That, paired with the fact that hospitals have an immense problem with overcrowding, is one of the challenges of the healthcare system for the next decade, if not longer.” He is director of emergency medicine research at Christchurch Hospital in Christchurch, New Zealand.
The Details of ASPECT
Than and his colleagues at 14 emergency departments in nine countries in the Asia-Pacific region launched the Asia-Pacific Evaluation of Chest Pain Trial (ASPECT) to assess whether a pre-defined protocol could identify patients presenting to the emergency department with chest pain who would be at low risk of harm if they were discharged early. The protocol’s POC panel consisted of cardiac troponin I (cTn), creatine kinase MB (CK-MB), and myoglobin. Researchers combined biomarker results with the Thrombolysis in Myocardial Infarction (TIMI) risk score and electrocardiograph (ECG) to establish patient risk.
ASPECT involved 3,582 consecutive adult patients who reported at least 5 minutes of chest pain suggestive of ACS. Patients received normal care, and attending physicians had access to central lab cTn results but were blinded to TIMI score and results from the POC panel, samples for which were drawn at admission and after 2-hours. The researchers combined the ASPECT protocol with medical records and telephone follow-up to determine the study’s primary endpoint, major adverse cardiac events within 30 days after initial presentation.
The POC panel results were considered positive when cTn was ≥0.05µg/L, CK MB was ≥4.3 µg/L or had an increase ≥1.6 µg/L within 2 hours, or myoglobin was ≥108 µg/L or increased ≥25% within 2 hours. Patients were deemed low risk if they had a TIMI score of 0, no new ischemic changes on ECG, and normal results from the POC biomarker panel, at both admission and 2-hours. The researchers found that 9.8% of patients were at low risk and would have been eligible for early discharge. Following release from the hospital, a major cardiac event occurred in three (0.9%) low-risk patients, giving the protocol a sensitivity of 99.3%, specificity of 11%, and negative predictive value of 99.1%.
Practice Changes Desired but Difficult
Observers agreed that a system which would lead to as many as 10% of suspected ACS patients being discharged quickly and safely would have considerable merit. “There’s overcrowding in emergency departments in the U.S. and world wide, so even a nine or 10 percent discharge rate that wasn’t there before would make a big difference in work flow,” said Alan Wu, PhD, director of clinical chemistry and toxicology at the University of California-San Francisco.
However, Wu was not alone in cautioning that as intriguing as the ASPECT results may be, a 2-hour assessment protocol is unlikely to be adopted in practice anytime soon. “It’s not a simple thing to change protocols. No single study would be the driving force to elicit such a change, but this will contribute to it,” he said. “Every institution would have to look at its own resources, needs, and turnaround times, and determine if it would be comfortable with a two-hour rule-out. A lot of hospitals in the U.S. will not be ready to adopt this today, especially where the medicolegal aspects are so different and we don’t have the luxury of missing an MI.”
Hospitals worldwide adhere to guidelines on the universal definition of MI issued in 2007 by the European Society of Cardiology, American College of Cardiology Foundation, American Heart Association, and World Heart Foundation. These guidelines call for a cTn measurement exceeding the 99th percentile of a normal reference population with a coefficient of variation ≤10% as an element of diagnosing MI, along with at least one additional criterion: symptoms or ECG changes indicative of ischemia; development of pathological Q waves in the ECG; or imaging evidence of new loss of viable myocardium or regional wall motion abnormality. The guidelines also note the importance of rising and/or falling cTn values in discerning MI. Measurements should be taken at the time of first assessment and 6–9 hours later to detect any pattern. CK-MB by mass assay is an acceptable alternative when cTn values are not available, according to the guidelines.
Guidelines for Managing Acute Coronary Syndrome Patients
Key professional organizations have published guidelines on the use and interpretation of cardiac biomarkers in ACS.
ACC/AHA 2007 Guidelines for the Management of Patients with Unstable Angina/Non-ST-Elevation Myocardial Infarction, J Am Coll Cardiol 2007;50:e1–157.
Clinical Policy: Critical Issues in the Evaluation and Management of Adult Patients with Non-ST-Segment Elevation Acute Coronary Syndromes, Annals of Emergency Medicine 2006;48:270–301.
NACB Laboratory Medicine Practice Guidelines for Utilization of Biochemical Markers in Acute Coronary Syndromes and Heart Failure, Clin Chem 2007;53:2086–2096.
Testing of Low-Risk Patients Presenting to the Emergency Department with Chest Pain: A Scientific Statement from the American Heart Association, Circulation 2010;122;1756–76.
Universal Definition of Myocardial Infarction, J Am Coll Cardiol 2007;50:2173–2195.
Cutting Down Assessment Times
Healthcare systems have adopted various strategies to adhere to the universal definition of MI while not keeping chest pain patients in emergency department beds per se. For example, many have chest pain or observation units where they transfer suspected ACS patients for continued work up. However, depending on how the units are staffed and where they’re located, they may still demand attention and resources from the emergency department for 6–8 hours, and in some instances, up to 12 hours.
Widespread adoption of the universal definition of MI and use of strategies like observation units reflect how far the fields of emergency medicine and cardiology have come over the past three decades in discerning MI from other sources of chest pain and moving towards shorter, more efficient assessment processes, according to Ezra Amsterdam, MD, associate chief of cardiovascular medicine at UC Davis Medical Center in Sacramento, Calif. “There’s been an evolution over the past 30 years. We’ve gone from admitting every adult patient with chest pain, and putting them through these rule-out procedures with serial cardiac enzymes and ECGs over several days until you finally decided it was OK to discharge the patient. That’s accelerated into protocols where a majority of patients are not admitted, and it’s done safely,” he explained. “So the ASPECT protocol is not new per se. What’s new is their systematic approach to a two-hour assessment.”
An Objective Method
Such a systematic approach to identifying patients at low risk has been a missing ingredient in emergency medicine, according to W. Frank Peacock, MD, vice chair of emergency medicine at the Cleveland Clinic Foundation. “Chest pain is the emergency department’s first or second most common presentation. The likelihood of an emergency doctor having a bad outcome in patients with chest pain occurs in the first five years of practice, so that tells you it’s really subjective,” he said. “What Martin did was to validate an accelerated diagnostic protocol with completely objective measures. There’s nothing subjective about it, and that’s the advantage.” Peacock helped analyze data and write the ASPECT report, but no Cleveland Clinic patients were part of the study.
Than added that demonstrating the value of a structured risk assessment was a key objective of ASPECT. “A lot of literature shows that assessments of pre-test probabilities are done extremely poorly by clinicians,” he observed. “In general, we tend to overestimate the risks because we’re concerned about the possibility of getting it wrong. We also tend not to agree with each other in assessing risk. That’s why we thought some sort of structured risk assessment was important.”
The Role of High-Sensitivity cTn Assays
While experts agreed that the ASPECT protocol broke ground by laying out objective assessment criteria for ACS, several observers questioned the relevancy of a POC panel employing CK-MB, myoglobin, and a contemporary—but not high-sensitivity—cTnI assay, given that both high-sensitivity cTnT and cTnI assays with limits of detection 10 times lower than conventional assays are about to hit the U.S. market (CLN Feb 2011). In addition, the cTnI assay used in ASPECT, the Alere Triage CardioProfiler, is not approved for use in the U.S.
“There’s no question that ASPECT adds to the argument that myoglobin and CK-MB are no longer necessary,” said Wu. “In my opinion that conclusion was reached years ago, even before the advent of the current generation of troponin assays, but this reaffirms even more so that labs should move away from the former two.”
Than explained that in his own clinical practice, he has not used myoglobin or CK-MB for at least a decade, but that the ASPECT investigators specifically wanted to evaluate a POC biomarker panel. “We were interested in the point-of-care concept, but we also felt that by using a slightly inferior troponin assay, one can say you can do this with any troponin assay you like—either point-of-care or central lab—because even a less-sensitive assay works just fine,” he explained. “The main message of the paper was that the combination of this risk/pre-test probability tool, with ECG and troponin or combinations of biomarkers can be used for safe, early discharge of some patients.
Whether the new high-sensitivity cTn assays will hinder or harm efforts to speed emergency department assessment of chest pain patients remains unclear. Some observers, like Michael Kontos, MD, argued that the assays will help quickly identify the lowest of low-risk patients but contribute in other ways to slower emergency department processing times. “You’ll have the really low-risk cohort where the troponin will be so sensitive that once you do the test and clinical assessment, you’ll essentially be able to exclude an acute coronary syndrome,” he said. “But then you’ll have a more intermediate risk group that will have detectible troponin levels—not necessarily with serial changes or highly suggestive of acute coronary syndrome—but they clearly need some sort of further evaluation.” Kontos is associate professor of internal medicine at Virginia Commonwealth University’s Pauley Heart Center in Richmond.
While high-sensitivity cTnI and cTnT assays might pose challenges in understanding just what very low levels of circulating cTn mean, Than suggested that when used with other components of the ASPECT protocol, the tests still would bring clarity and speed to emergency department-based chest pain evaluations. “There are far more patients who need to be ruled-out than ruled-in, and the way I see the high-sensitivity troponin assays working is that you’ll be able to set a slightly higher pretest probability to your risk score because the assay will be more sensitive,” he said. “This will enable you to incorporate a broader group of patients in a potential early rule-out group. There needs to be a greater awareness of rule-out and the benefits it can have on the healthcare system.”
A Glimpse of the Future
Most experts agreed that the high-sensitivity cTnT and cTnI assays eventually would lead the way towards shorter chest pain assessment protocols. “Conceptually, this a great model to consider, and it’s a good first step in thinking about whether assays will be able to differentiate chest pain patients this early in the process. The high-sensitivity troponin assays will narrow the window,” predicted Fred Apple, PhD, professor of laboratory medicine and pathology at the University of Minnesota School of Medicine, and medical director of clinical laboratories at Hennepin County Medical Center in Minneapolis.
However, Apple cautioned that considerably more research would be needed before 2-hour chest pain work-ups become standard-of-care. “We’ll need a wealth of more information before we start changing practice patterns. The high-sensitivity troponin assays will need to be looked at for baseline and two hours for every person who walks in the emergency department with suspected acute coronary syndrome, and determine whether an absolute value or delta change percent is best for patient triage. This is not ready for primetime just yet.”
Research efforts that might transform chest pain assessment standards already are underway. For example, a team of British researchers recently reported that patients with suspected MI who were tested at presentation and after 90 minutes with a POC biomarker panel consisting of cTn, CK-MB, and myoglobin had shorter median, but not mean, lengths of stay in the emergency department and were discharged more frequently without inpatient admission (Heart 2011;97:190–196).
In addition, Amsterdam and his colleagues at UC Davis are preparing to publish results from an accelerated assessment process for very low risk patients. “We’ve been investigating a protocol that involves a clinical assessment, biomarkers, and ECG with a two-to-four hour period,” he said. “It’s a minority of our population, but we’ve found the patients we’re doing this with to be very safe with no events at 30-days.”
Than also has initiated a randomized controlled trial using the ASPECT protocol but with a central lab-based cTn assay. “I wouldn’t expect anyone to change their practice necessarily based on the ASPECT study, but I might expect them to take more interest when results from this second study become available,” he said. “We expect that we’ll demonstrate that you can send 15 to 20 percent of patients home early and safely, with an economic benefit in terms of bed days saved.”