Steve Goodacre PhD FCEM
School of Health and Related Research,
University of Sheffield, Sheffield, UK
Health services across the world face rising numbers of emergency admissions. A report by the Nuffield Trust(1) revealed that the number of emergency admissions in England rose by 11.8% between 2004/05 and 2008/09, resulting in around 1.35 million extra admissions at an estimated additional cost of £330 million per year. The report also found a dramatic increase in the number of short-stay admissions of one day or less, from 42% of all emergency admissions in 2004/5 to 49% in 2008/09. This suggests that rising admissions are not simply explained by an ageing population with more morbidity but may be due to a change in decision-making thresholds, with lower risk patients being admitted to hospital for short stays. If these additional admissions provide no benefit, then patients are being exposed to unnecessary inconvenience and risks such as hospital-acquired infection, and health care resources are being wasted.
Chest pain is responsible for around a quarter of emergency medical admissions.(2) The development of time-critical treatments for myocardial infarction has led to public education campaigns encouraging patients with acute chest pain to attend hospital as an emergency. However, due to the limited early sensitivity of troponin assays for myocardial infarction, guidelines recommend delaying testing until
10–12 hours after symptom onset.(3,4) Because patients typically present around two hours after symptom onset, this means that many are admitted to hospital for delayed troponin testing.
Is there a solution?
Point-of-care multiple biomarker testing offers a potential solution to this problem. Point-of-care testing allows biomarker results to be rapidly available, while multiple biomarkers improve early sensitivity. Patients with a negative biomarker assessment can potentially be discharged home within a few hours of hospital arrival, thus reducing admission rates.
Most multiple marker strategies involve a combination of troponin, creatine kinase MB (CKMB) and myoglobin measured at presentation and repeated between 60 and 180 minutes later. These markers have been combined into a single point-of-care assessment by the Biosite Triage analyser.(5) However, the definition of a positive assessment varies, with the gradient rise of myoglobin or CKMB often being used instead of absolute measurements to optimise sensitivity without sacrificing specificity.
As shown in Table 1, cohort studies have evaluated the diagnostic accuracy of point-of-care multiple marker combinations. Most report high sensitivity and specificity, although interpretation of these findings is limited by variation in the way biomarker combinations are used and outcomes defined. Other studies have shown that a point-of-care multiple marker strategy allows risk stratification for subsequent adverse events.(11,12)
Most recently, the ASPECT study(13) evaluated an accelerated diagnostic protocol for patients with chest pain based upon the Thrombolysis in Myocardial Infarction (TIMI) score, electrocardiograph and point-of-care biomarker panel of troponin, CKMB and myoglobin at presentation and two hours later. The protocol identified major adverse events over the following 30 days (421/3582; 11.8%) with a sensitivity of 99.3% (95% confidence interval, 97.9–99.8) and a specificity of 11.0% (10.0–12.2).
These findings suggest that the combination of troponin, CKMB and myoglobin can rule out myocardial infarction and identify patients with a low risk of adverse outcome, but evidence of a change in patient management is required before general implementation. Newby et al(11) demonstrated identification of myocardial infarction earlier than laboratory testing, and Carragher et al(8) showed expedited decision-making with turnaround times reduced by 55%, but it was not clear whether these led to meaningful changes in patient care. Ng et al(7) compared management with the panel to previous practice and showed a 40% reduction in coronary care unit admissions. However, historically controlled studies tend to over-estimate the effects of intervention.
Randomised trials have compared use of point-of-care and laboratory troponin assays.(14,15) Renaud et al(14) showed that point-of-care troponin testing in an emergency department reduced time to offering anti-ischaemic therapy and physician notification of troponin results but did not change emergency department length of stay or patient outcomes. Ryan et al(15) evaluated point-of-care troponin testing in four emergency departments and found that the effect varied between settings, with length of stay in the emergency department being increased in one hospital and decreased in another. These studies show that using point-of-care assays instead of laboratory assays can change patient management, but they do not specifically evaluate the potential for multiple point-of-care biomarker assessment to reduce hospital admissions.
The RATPAC trial (Randomised Assessment of Treatment using Panel Assay of Cardiac markers)(16) evaluated the point-of-care combination of troponin, CKMB and myoglobin at presentation and 90 minutes. The trial took place in six hospitals in England and Scotland. Patients with acute chest pain due to suspected but not proven myocardial infarction and no other serious illness were randomised to point-of-care assessment or standard care without the point-of-care tests.
The RATPAC trial showed that point-of-care assessment increased the proportion of patients successfully discharged home: 32% in the point-of-care group versus 13% in the standard care group (odds ratio 3.81; 95% confidence interval, 3.01 to 4.82; p<0.001).(17) Figure 1 shows the proportion of patients in each study group in hospital as a function of time from arrival. The proportion of patients in hospital was lower in the point-of-care group up to 24 hours after arrival, from which point the proportions in the two groups were similar. Thus, it is evident that the admissions avoided by using point-of-care testing were less than 24 hours duration.
Point-of-care testing reduced the median length of initial hospital stay (8.8 versus 14.2 hours; p<0.001) but not mean length of the initial stay (29.6 versus 31.7 hours; p=0.462) or mean in-patient days over follow-up (1.8 versus 1.7 days; p=0.815). These findings were influenced by a small proportion of patients with long hospital stays and suggest that the reduction in hospital admissions associated with point-of-care testing did not lead to a reduction in hospital bed use.
Point-of-care testing was associated with an increased proportion of patients being admitted to coronary care (4.2 versus 2.8%; p=0.041) and a non-statistically significant increase in coronary interventions (4.2 versus 3.4%; p=0.061). As a result, point-of-care assessment did not lead to reduced health and social care costs and may even have increased costs (mean cost per patient £1216.(18) versus £1008.94; p=0.056).(18)
In summary, the RATPAC trial showed that point-of-care assessment reduced hospital admissions, but this did not lead to reduced bed use or costs and may even have increased costs. This conclusion applies broadly across the UK National Health Service, but care should be taken in extrapolating to other health care settings or specific hospitals. Analysis across the six RATPAC hospitals(19) showed that the odds ratio for the primary outcome varied from 0.12 to 11.07, with significant heterogeneity between hospitals (p<0.001), while the mean cost per patient for the intervention group ranged from being £214.49 less than the control group to £646.57 more expensive, with weak evidence of heterogeneity (p=0.0803). These findings suggest that point-of-care assessment could be more useful in certain settings. Moreover, the potential for reducing hospital admissions and saving costs could be greater in countries that have higher admission rates than the United Kingdom, although this needs confirmation in well-designed randomised trials. One such trial is currently in progress in Christchurch, New Zealand.
Recent studies have evaluated the sensitivity of modern high-sensitivity troponin assays when patients present to hospital.(20,21) These studies suggest that modern assays have sensitivity for myocardial infarction of around 90% when the conventional threshold of the 99th percentile is used for decision-making. Using a lower threshold such as the limit of detection could improve sensitivity further but with a potentially unacceptable loss of specificity. Measuring the absolute gradient rise of troponin can optimise early sensitivity and specificity.(22) The RATPAC trial used a sensitive troponin assay, and secondary analysis of the RATPAC data showed that CKMB and myoglobin did not add diagnostic value to that provided by troponin.(23) These studies suggest that point-of-care assessment with troponin alone measured at baseline and 90 minutes could be more effective in reducing admissions than a multiple marker approach. However, not all point-of-care analysers use state-of-the-art troponin assays with appropriate validation.
Ultimately, it could be argued that we are looking at the problem of diagnosis the wrong way round, and this is driving increased hospital admissions. We know that even a 12-hour troponin does not rule out coronary heart disease and eliminate any risk of subsequent adverse events, so the purpose of hospital assessment is to identify those at higher risk who will benefit from hospital treatment. Most of those at higher risk will be identified by an admission troponin. Rational clinical practice should demand evidence that admission for delayed troponin testing for those with a negative troponin at presentation provides benefits at an acceptable cost before inflicting this intervention upon patients and incurring costs for the health service.
Conclusions
Point-of-care assessment with multiple markers reduces hospital admissions with chest pain, but this does not appear to translate into reduced bed use or cost savings. Indeed, it may even increase costs.
Using a modern high sensitivity troponin assay on its own is probably as sensitive and more specific than a multiple marker approach, although not all point-of-care analysers use state of the art troponin assays.
Reducing hospital admissions may be best achieved by questioning the value of admission and demanding evidence of benefit before admission is recommended in guidelines.
References
- Blunt I, Bardsley M, Dixon J. Trends in hospital admissions in England 2004–2009. The Nuffield Trust, 2010.
- Goodacre S et al. The health care burden of acute chest pain. Heart 2005;91:229–30.
- National Institute for Health and Clinical Excellence. Chest pain of recent onset: Assessment and diagnosis of recent onset chest pain or discomfort of suspected cardiac origin. www.nice.org.uk/guidance/CG95
- Bertrand ME et al. Task Force on the Management of Acute Coronary Syndromes of the European Society for Cardiology. Management of acute coronary syndromes in patients presenting without ST elevation. Eur Heart J 2002;23:1809–40.
- Apple FS et al. Simultaneous rapid measurement of whole blood myoglobin, creatinine kinase MB and cardiac troponin I by the triage cardiac panel for detection of myocardial infarction. Clin Chem 1999;45:199–205.
- McCord J et al. Ninety-minute exclusion of acute myocardial infarction by use of quantitative point-of-care testing of myoglobin and troponin I. Circulation 2001;104:1483–8.
- Ng SM et al. Ninety-minute accelerated critical pathway for chest pain evaluation. Am J Cardiol 2001;88:611–7.
- Caragher TE et al. Evaluation of quantitative cardiac biomarker point of care testing in the emergency department. J Emerg Med 2002;22:1–7.
- Rathore S et al. Is it safe to discharge patients from accident and emergency using a rapid point of care Triple Cardiac Marker test to rule out acute coronary syndrome in low to intermediate risk patients presenting with chest pain? Eur J Int Med 2008;19:537–40.
- Straface AL et al. A rapid point-of-care cardiac marker testing strategy facilitates the rapid diagnosis and management of chest pain patients in the emergency department. Am J Clin Pathol 2008;129:788–95.
- Newby LK et al. Bedside multimarker testing for risk stratification in chest pain units: The CHECKMATE Study. Circulation 2001;103:1832–7.
- Hamilton AJ et al. Risk stratification of chest pain patients in the emergency department by a nurse utilizing a point of care protocol. Eur J Emerg Med 2008;15:9–15.
- Than M et al. A 2-h diagnostic protocol to assess patients with chest pain symptoms in the Asia-Pacific region (ASPECT): A prospective observational validation study. Lancet 2011;377:1077–84.
- Renaud B et al. Impact of point-of-care testing in the emergency department: Evaluation and treatment of patients with suspected acute coronary syndromes. Acad Emerg Med 2008;15:216–24.
- Ryan RJ et al. A multicenter randomized controlled trial comparing central laboratory and point-of-care cardiac marker testing strategies: The Disposition Impacted by Serial Point of Care Markers in Acute Coronary Syndromes (DISPO-ACS) trial. Ann Emerg Med 2009;53:321–8.
- Goodacre S et al. The RATPAC (Randomised Assessment of Treatment using Panel Assay of Cardiac markers) trial: A randomised controlled trial of point-of-care cardiac markers in the emergency department. Health Technol Assess 2011;15(23):1–108.
- Goodacre SW et al. On behalf of the RATPAC research team. The RATPAC Trial (Randomised Assessment of Treatment using Panel Assay of Cardiac markers): A randomised controlled trial of point-of-care cardiac markers in the emergency department. Heart 2011;97:190–6.
- Fitzgerald P et al. Cost-effectiveness of point-of-care biomarker assessment for suspected myocardial infarction: The RATPAC Trial (Randomised Assessment of Treatment using Panel Assay of Cardiac markers). Acad Emerg Med 2011;18:488–95.
- Bradburn M et al. Inter-hospital variation in the RATPAC Trial (Randomised Assessment of Treatment using Panel Assay of Cardiac markers). Emerg Med J 2012;29:233–8.
- Reichlin et al. Early diagnosis of myocardial infarction with sensitive cardiac troponin assays. New Engl J Med 2009;361:858–67.
- Keller et al. Sensitive troponin I assay in early diagnosis of acute myocardial infarction New Engl J Med 2009;361:868–72.
- Reichlin T et al. Utility of absolute and relative changes in cardiac troponin concentrations in the early diagnosis of myocardial infarction. Circulation 2011;124:136–45.
- Collinson P, Goodacre S. Rapid diagnostic protocol for patients with chest pain (letter). Lancet 2011;378:397.
