Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Undifferentiated chest pain is a presenting complaint facing the emergency medicine clinician with ever increasing frequency.1 Increasing public awareness regarding ischaemic heart disease has helped contribute towards understandable increasing concern. The net result is that patients hold a lower threshold to present to the emergency department (ED) for assessment.
With these patients holding a lower pretest probability for acute coronary syndrome, the art of emergency medicine is being able to risk-stratify patients into those requiring admission for further assessment for a possible acute coronary syndrome and discharging home those patients whose likelihood of the disease is under the test threshold. Setting your threshold too high for investigation leads to unacceptable low pick up of cases and a missed opportunity to treat a disease with a significant morbidity and mortality; set your threshold too low for investigation and not only do you exposure patients to the elevated risk of false positives and unnecessary and potentially harmful treatment but you also increase the burden of over investigation upon an acute healthcare system already bursting at the seams.
It is well documented that clinicians are not able to accurately estimate the pretest probability for acute coronary syndrome accurately, with a persistent overestimation of patients risk.2 Classical teaching regarding risk assessment for potential acute coronary syndrome and many current guidelines place the emphasis upon the importance of risk factors for ischaemic heart disease (IHD), which confer a chronic risk of developing the disease, but these have been inappropriately used for the assessment of acute coronary syndrome in the ED.3 ,4
Thrombolysis in Myocardial Infarction (TIMI) and Global Registry for Acute Coronary Events (GRACE) scoring for acute coronary syndrome are among the most frequently used in the UK. The GRACE score gives estimates for 6-month morbidity and mortality.5 Although useful, GRACE's longer period of prognostication is too long to reflect the likely follow-up time for investigation following discharge from ED for IHD. TIMI risk calculation requires background information to be available; if the patient has had an angiogram, the clinician needs to ascertain whether there was >50% stenosis. This information may not always be readily available affecting the validity of the tool.
The Manchester Acute Coronary Syndromes (MACS) trial, conducted by Body et al,6 derives and validates a new decision rule for suspected cardiac chest pain. They acknowledge that the current gold standard of investigation often necessitates a hospital admission to rule out the diagnosis of acute coronary syndrome and a cohort of false negatives. The study successfully demonstrates the ability to identify a significant cohort of patients who can be classified as ‘very low risk’, which was shown to infer a risk of 0.0% of AMIs within the next 30 days and a risk of missed acute coronary events (MACE) of 1.6% who may be deemed low enough risk for safe discharge.6 More than just identifying a dichotomous admit/discharge category, it enabled patients to be placed into one of several quantitative categories of risk for AMI/MACE.
The factors incorporated into the MACS decision tool do hold benefits over the GRACE and TIMI scores as it prognosticates over a more appropriate time frame of 30 days and can be accurately evaluated by information not requiring previous patient investigations. Sadly this is not a risk stratification tool that is ready for the prime time just yet. One of the biomarkers, heart-type fatty acid binding protein will not be available in many UK biochemistry laboratories but may well be in years to come.
The HEART score, as mentioned in this paper’s discussion, is a pre-existing risk stratification tool for chest pain that incorporates features from the history, ECG, past medical history and a serum troponin result.7 The obvious benefit to this tool being that it uses a biomarker that is readily available in hospitals already and could be implemented currently. Again this score splits patients into those at a low risk (<1.6% of MACE) in the next 60 days. It also suggests a management strategy dependent upon the level of risk calculated by the tool. As the authors suggest, the MACS score will require evaluation in comparison to the HEART score.
As clinicians we like risk assessment tools as they are explicit in telling us what to do with our patients. They remove concern over where our test threshold and discharge threshold lie and enable us to deliver a conclusion over further care. What they do not do is consider the patient’s wishes. As patients our acceptance of medical risk varies from one individual to the next. We all place a varying degree of importance upon different aspects of our lives and as such prefer to be investigated in a way tailored to our beliefs. In an era of patient-centred care and patient choice, a risk assessment tool may provide a much needed quantification of individual patients risk for a significant cardiac event. This allows them to fully consent for further investigation or indeed against further inpatient work up, by allowing them to truly understand what they have to gain through inpatient investigation. Would a patient really wish to be admitted to hospital for further investigation and treatment solely because they have a 1% greater chance of a significant cardiac event than the patient we think is safe for discharge? When is low-risk chest pain acceptable risk chest pain to us and to our patients?
The MACS tool certainly provides a way to quantify risk using potential biomarkers of the future and will benefit from evaluation against other similar risk assessment tools. Perhaps now is also the time to seriously evaluate the false positives picked up by current practice and weigh it against the hidden risk that we add, the risk of over investigation and treatment.
Contributors SJB-M was invited to write the editorial. Both SJB-M and SL wrote and reviewed the script.
Competing interests None.
Provenance and peer review Commissioned; internally peer reviewed.