Improving the risk stratification, diagnosis and classification of patients with suspected myocardial infarction

Abstract

Myocardial infarction is a leading cause of morbidity and mortality worldwide. The purpose of this thesis was to develop strategies for the assessment of patients with suspected myocardial infarction using a high-sensitivity cardiac troponin I assay, and to evaluate the relationship between the aetiology of myocardial infarction and long term clinical outcomes to identify opportunities to modify outcomes. In the United Kingdom, approximately 1 million patients present to hospital with chest pain each year and are assessed for suspected myocardial infarction, yet fewer than 20% of patients receive this diagnosis. Prior clinical standards mandated the admission of patients for serial cardiac troponin testing to identify myocardial necrosis and determine if myocardial infarction had occurred. However, new high-sensitivity assays offer a magnitude improvement in diagnostic precision, and as such provide a novel approach to diagnose or exclude myocardial infarction at an earlier stage. In our first study, I evaluate the performance of a high-sensitivity cardiac troponin I assay as a risk stratification tool in patients with suspected acute coronary syndrome.

A systematic review and individual patient-level data meta-analysis was performed, including prospective studies measuring high-sensitivity cardiac troponin I in patients with suspected acute coronary syndrome, where the diagnosis was adjudicated according to the universal definition of myocardial infarction. The primary outcome was myocardial infarction or cardiac death during the index hospitalization or at 30 days. Meta-estimates for primary and secondary outcomes were derived using a binomial-normal random effects model. Performance was evaluated in subgroups and across a range of troponin concentrations (2-16 ng/L) using individual patient data.

A total of 22,457 patients were included in the meta-analysis (age 62 [15.5] years; n=9,329 (41.5%) women), of whom 2,786 (12.4%) experienced myocardial infarction or cardiac death at 30 days. Cardiac troponin I concentrations were <5 ng/L at presentation in 11,012 (49%) patients, with a negative predictive value of 99.5% (95% confidence interval [CI] 99.3-99.6) for myocardial infarction or cardiac death at 30 days. Lower thresholds did not improve safety, but did significantly reduce the proportion identified as low risk.

This threshold of 5 ng/L formed the basis for the development of a diagnostic pathway for patients with suspected acute coronary syndrome. In a cohort study of 1,218 patients with suspected acute coronary syndrome who underwent high-sensitivity cardiac troponin I measurement at presentation, 3 and 6 or 12 hours, I derived and validated a novel pathway (rule out myocardial infarction if <5 ng/L at presentation, or change <3 ng/L and <99th centile at 3 hours), and compared this with the established European Society of Cardiology 3-hour pathway (rule out myocardial infarction if <99th centile at presentation, or at 3 hours if symptoms <6 hours). The primary outcome was a comparison of the negative predictive value (NPV) of both pathways for myocardial infarction or cardiac death at 30 days. The primary outcome was evaluated in pre-specified subgroups stratified by age, gender, time of symptom onset and known ischaemic heart disease.

In those <99th centile at presentation, the ESC pathway ruled out myocardial infarction in 28.1% (342/1,218) and 78.9% (961/1,218) at presentation and 3 hours respectively, missing 18 index and two 30-day events (NPV 97.9%, 95% confidence intervals [CI] 96.9-98.7%). The novel pathway ruled out 40.7% (496/1,218) and 74.2% (904/1,218) at presentation and 3 hours, missing two index and two 30-day events (NPV 99.5%, 95% CI 99.0-99.9%; P<0.001 for comparison). The NPV of the novel pathway was greater than the ESC pathway overall (P<0.001), and in all subgroups including those presenting early or known to have ischaemic heart disease.

There are a number of additional approaches for the rule out of myocardial infarction.

Clinical risk scores apply conventional risk factors to estimate the probability of myocardial infarction. The most widely implemented scores, HEART, EDACS, GRACE and TIMI, have been extensively validated when used alongside contemporary troponin assays, however, their impact on pathways applying high-sensitivity cardiac troponin testing is less clear. In 1,935 patients with suspected acute coronary syndrome, I evaluated the safety and efficacy of our novel pathway or the European Society of Cardiology 3-hour pathway alone, or in conjunction with low-risk TIMI (0 or 1), GRACE (≤108), EDACS (<16) or HEART (≤3) scores. Myocardial infarction or cardiac death at 30-days occurred in 14.3% (276/1,935). The ESC pathway ruled out 70% with 27 missed events giving a negative predictive value (NPV) of 97.9% (95% confidence interval [CI], 97.1 to 98.6%). Addition of a HEART score ≤3 reduced the proportion ruled out by the ESC pathway to 25%, but improved the NPV to 99.7% (95%CI 99.0 to 100%, P<0.001). The novel pathway ruled out 65% with three missed events for a NPV of 99.7% (95%CI 99.4 to 99.9%). No risk score improved the NPV, but all reduced the proportion ruled out (24-47%, P<0.001 for all).

Whilst myocardial infarction due to atherosclerotic plaque rupture and thrombosis (type 1) is well described, the natural disease course of myocardial infarction due to oxygen supply-demand imbalance without atherothrombosis (type 2) is poorly understood. I aimed to define long-term outcomes and explore risk stratification in patients with type 2 myocardial infarction and myocardial injury. Consecutive patients (n=2,122) with elevated cardiac troponin I concentrations (≥0.05 μg/L) were identified at a tertiary cardiac centre. All diagnoses were adjudicated as per the Universal Definition of Myocardial Infarction. The primary outcome was all-cause death.

Secondary outcomes included major adverse cardiovascular events (MACE; non-fatal myocardial infarction or cardiovascular death) and non-cardiovascular death. To explore competing risks, cause-specific hazard ratios were obtained using Cox regression models.

The adjudicated index diagnosis was type 1 or type 2 myocardial infarction or myocardial injury in 1,171 (55.2%), 429 (20.2%) and 522 (24.6%) patients, respectively. At five years, all-cause death rates were higher in those with type 2 myocardial infarction (62.5%) or myocardial injury (72.4%) compared with type 1 myocardial infarction (36.7%). The majority of excess deaths in those with type 2 myocardial infarction or myocardial injury were due to non-cardiovascular causes (HR 2.32, 95%CI 1.92-2.81, versus type 1 myocardial infarction). Despite this, the observed crude MACE rates were similar between groups (30.6% versus 32.6%), with differences apparent after adjustment for co-variates (HR 0.82, 95%CI 0.69-0.96). Coronary heart disease was an independent predictor of MACE in those with type 2 myocardial infarction or myocardial injury (HR 1.71, 95%CI 1.31-2.24). Patients with type 2 myocardial infarction were less likely to receive secondary prevention therapy, suggesting a treatment gap may exist and there may be potential to modify clinical outcomes.

A risk stratification threshold has been defined using high-sensitivity cardiac troponin I which identifies patients at very low risk of myocardial infarction or cardiac death.

A diagnostic pathway incorporating this risk stratification threshold appears safer than established guidelines which apply the 99th centile alone. The use of clinical risk scores does not appear to improve the safety of this approach, however, does significantly reduce efficacy. Overall, these findings demonstrate the potential of high-sensitivity cardiac troponin testing to improve the efficiency of the assessment of patients with suspected acute coronary syndrome without compromising patient safety.

The observations in those with myocardial injury and infarction have identified a phenotype of patients with type 2 myocardial infarction and coronary artery disease who are at increased cardiovascular risk, and who may benefit from targeted secondary prevention. The studies presented will inform the design of future clinical trials, and may inform international guidelines for the assessment of patients with suspected acute coronary syndrome.