Advertisement
Canadian Journal of Cardiology

Identifying very low risk patients for future myocardial infarction or death

Open AccessPublished:October 15, 2022DOI:https://doi.org/10.1016/j.cjca.2022.10.014
      High-sensitivity cardiac troponin (hs-cTn) testing has enabled shorter time intervals between serial measurements when assessing patients with possible acute coronary syndrome (ACS) with a single sample strategy proposed for early risk stratification. Specifically, a rapid rule-out for myocardial infarction (MI) is suitable for a pathway if the sensitivity for 30-day cardiac events is ≥99%. At the population level [n=131,095 emergency department (ED) patients], low hs-cTn results alone yielded sensitivities <99%. Test performance was improved with additional laboratory tests, with the combination of glucose, the estimated glomerular filtration rate (eGFR) and hs-cTnT or hs-cTnI yielding sensitivities >99.0% in patients ≥65 years.
      A recent clinical practice guideline from the American College of Cardiology/American Heart Association for the evaluation and diagnosis of acute chest pain lists different clinical decision pathways (CDPs) that may be used to further classify patients with suspected ACS into different risk categories.

      Gulati M, Levy PD, Mukherjee D, et al. 2021 AHA/ACC/ASE/CHEST/SAEM/SCCT/SCMR Guideline for the Evaluation and Diagnosis of Chest Pain: Executive Summary: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation 2021;144:e368-e454.c

      These guidelines place emphasis on time of symptom onset for CDPs that use the limit of detection or low cutoff for hs-cTn assays for rapid rule-out.

      Gulati M, Levy PD, Mukherjee D, et al. 2021 AHA/ACC/ASE/CHEST/SAEM/SCCT/SCMR Guideline for the Evaluation and Diagnosis of Chest Pain: Executive Summary: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation 2021;144:e368-e454.c

      However, from a clinical perspective, time of symptom onset may be misinterpreted and electrocardiogram (ECG) changes can be nonspecific in many ED patients.
      From the laboratory perspective, gaps affecting hs-cTn interpretation on what constitutes an undetectable or low hs-cTn concentration exist. For instance, laboratories may select different lower limits for reporting using either the instruction for use/package insert from the respective test manufacturer or deriving their own lower limit. Other times, a regulatory body (e.g., FDA in the United States) will mandate that the lower limit of reporting be set to a specific metric for hs-cTn assays (i.e., the limit of quantification). Collectively, there is no universal laboratory standardization with respect to what lower limit should be used, and as such is a source of variation and misclassification. Our objective was to assess laboratory testing alone in a large ED population using low hs-cTn cutoffs in isolation and in combination with other common laboratory tests for future risk stratification.

      METHODS

      Participants

      Patients who presented to the ED with acute chest pain and evaluated for ACS between April 2013 and December 2020 were identified at hospitals in the province of Ontario via databases (i.e., the discharge abstract database for hospitalizations, the national ambulatory care reporting system [NACRS] for ED visits, the registered persons database for demographics and death, the Ontario Health Insurance Plan [OHIP], and the Ontario Laboratory Information System [OLIS] for laboratory data). During this time period, the same hs-cTnT assay generation was used (Roche Diagnostics) with other hs-cTnI assays gradually being approved by Health Canada (Abbott Diagnostics, Beckman Coulter, Siemens Healthcare Diagnostics) and being used by hospital laboratories within Ontario. These datasets were linked using unique encoded identifiers and analyzed at ICES, formerly known as the Institute for Clinical Evaluative Sciences.
      • Lau G.
      • Koh M.
      • Kavsak P.A.
      • et al.
      Clinical outcomes for chest pain patients discharged home from emergency departments using high-sensitivity versus conventional cardiac troponin assays.
      Patients were included if they presented to ED with chest pain during the study timeline (International Classification of Disease 10th Revision: R071-R074, I200, I201, I208, and I209) (n=1,708,034). Patients were excluded if they were admitted to hospital or presented to a pediatric or mental health hospital (n=82,883), aged outside 40-105 years (n=480,674), not enrolled in OHIP (n=2,340), no ECG performed (n=130,524), and did not have hs-cTn reported and other laboratory tests (e.g., creatinine, glucose) (n=880,518) that are useful for risk stratification (n=131,095 patients for final population) (Supplemental Figure S1 for Flow Diagram).
      • Lau G.
      • Koh M.
      • Kavsak P.A.
      • et al.
      Clinical outcomes for chest pain patients discharged home from emergency departments using high-sensitivity versus conventional cardiac troponin assays.
      The first ED visit with the necessary lab results was used as the index to assess the primary outcome, which was defined as all-cause mortality or MI after ED discharge up to 365-days.

      Measurement

      Two single blood draw approaches were evaluated for classifying low-risk patients. First, a low hs-cTn cutoff alone, a low hs-cTnT was defined as <6 ng/L (lowest reportable concentration in the United States) and a low hs-cTnI defined as <5 ng/L; a concentration that has been used for identifying low-risk patients with several different hs-cTnI assays (Approach 1).
      • Kavsak P.A.
      • Cerasuolo J.O.
      • Ko D.T.
      • et al.
      Using the clinical chemistry score in the emergency department to detect adverse cardiac events: a diagnostic accuracy study.
      ,
      • Westwood M.E.
      • Armstrong N.
      • Worthy G.
      • et al.
      Optimizing the Use of High-Sensitivity Troponin Assays for the Early Rule-out of Myocardial Infarction in Patients Presenting with Chest Pain: A Systematic Review.
      A second method was utilized where patients with normal kidney function (eGFR ≥90mL/min/1.73 m2 with eGFR calculated via the CKD-EPI equation, normal glycemia (glucose <5.6 mmol/L) and normal hs-cTn (hs-cTnT <8 ng/L or hs-cTnI <4 ng/L) were classified as low-risk (i.e., low-risk defined by the clinical chemistry score <1) (Approach 2).
      • Kavsak P.A.
      • Cerasuolo J.O.
      • Ko D.T.
      • et al.
      Using the clinical chemistry score in the emergency department to detect adverse cardiac events: a diagnostic accuracy study.

      Data analysis and Outcomes

      Descriptive (non-parametric) and diagnostic parameters (i.e., sensitivity and negative predictive value [NPV] with 95% confidence intervals; CIs) were calculated using the clinical chemistry score and the hs-cTn respective cutoffs in the overall population. Forest plots on important subgroups (i.e., sex, age) and the most common conditions/procedures (i.e., history of hypertension, diabetes, and previous stress test) for the sensitivity of the primary outcome at 30-days with both approaches was also performed. Analyses were performed using SAS 9.1.3 software (SAS Institute Inc, Cary, NC, USA) and MedCalc Statistical Software version 20.109 (MedCalc Software Ltd., Ostend, Belgium) for the primary outcome (death/MI). Ethics approval granted by the Hamilton Integrated Research Ethics Board (HiREB: 4717-D).

      RESULTS

      Of the 131,095 patients discharged home from the ED, 52.7% were evaluated with hs-cTnT as compared to 47.3% with hs-cTnI. The 30-day incidence of death/MI was <0.50% with NPV ≥99.5% for hs-cTn alone (Approach 1) and for the clinical chemistry score (Approach 2) (Table 1). However, the sensitivities were higher for Approach 2 (range: 97.3-98.4%) as compared to Approach 1 (range:53.9-92.7%) over the first year (Supplemental Table S1). Only Approach 2 yielded an incidence rate <1.00% at 180 and 365-days. Subgroup analyses at 30-days yielded sensitivity point estimates >99% only for Approach 2 (Figure 1). Here, using Approach 2 with hs-cTnT, the sensitivity for hs-cTnT was >99% for females, those 65 years and older or patients with a history of hypertension or diabetes.
      Table 1Specific characteristics and 30-day death/MI for patients discharged home and evaluated with hs-cTnT (n=69133) and hs-cTnI (n=61,962) in the province of Ontario from 2013 to 2020.
      Approach 1 with hs-cTnT n = 30,365 (44%)Approach 2 with hs-cTnT

      n= 9,107 (13%)
      Approach 1 with hs-cTnI n=37,182 (60%)Approach 2 with hs-cTnI n=6,005 (10%)
      Age (years), median (IQR)52 (46-60)49 (44-56)57 (49-67)49 (44-56)
      Female, n (%)20,169 (66.4%)5,833 (64.0%)19,923 (53.6%)3,306 (55.1%)
      History, n (%)
      Hypertension10,325 (34.0%)2,377 (26.1%)17,825 (47.9%)1,663 (27.7%)
      Stress Test10,277 (33.8%)2,717 (29.8%)14,648 (39.4%)1,808 (30.1%)
      Diabetes4,173 (13.7%)619 (6.8%)8,093 (21.8%)401 (6.7%)
      30-day death/MI, n (%)37 (0.12%)8 (0.09%)156 (0.42%)9 (0.15%)
      Sensitivity (95% CI)92.7%(90.1%-94.8%)98.4%(96.9%-99.3%)57.8%(52.6%-62.9%)97.6%(95.4%-98.9%)
      Negative Predictive Value (95% CI)99.9%(99.8%-99.9%)99.9%(99.8%-99.9%)99.6%(99.5%-99.6%)99.9%(99.7%-99.9%)
      Note: IQR = interquartile range; CI = confidence interval; MI = myocardial infarction
      Figure thumbnail gr1
      Figure 1Forest plot of the sensitivity for MI/death at 30 days for Approach 1 (hs-cTn alone) and Approach 2 (Clinical Chemistry Score) among different subgroups stratified by hs-cTnT and hs-cTnI testing.

      DISCUSSION

      Over 8 years at the population level in Ontario, patients discharged home from the ED had a 30-day death/MI incidence of <0.50%, using either a low hs-cTn cutoff alone or with inclusion of normal kidney function and normal glycemia. The overall sensitivities with either approach for death/MI was <99%, lower than the benchmark used in some guidelines.

      Gulati M, Levy PD, Mukherjee D, et al. 2021 AHA/ACC/ASE/CHEST/SAEM/SCCT/SCMR Guideline for the Evaluation and Diagnosis of Chest Pain: Executive Summary: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation 2021;144:e368-e454.c

      However, in patients 65 years and older, the sensitivity with Approach 2 (the clinical chemistry score) was >99% using either hs-cTnT or hs-cTnI. Also, Approach 2 yielded higher and similar diagnostic estimates using all hs-cTn assays (i.e., sensitivity >97%) over 365 days and exceeded another proposed sensitivity benchmark for rule-out of MI in chest pain patients from a large systematic review undertaken for the National Institute for Health and Care Excellence guidance in the United Kingdom.
      • Westwood M.E.
      • Armstrong N.
      • Worthy G.
      • et al.
      Optimizing the Use of High-Sensitivity Troponin Assays for the Early Rule-out of Myocardial Infarction in Patients Presenting with Chest Pain: A Systematic Review.

      Limitations

      There are limitations to this large population-based study. First, laboratory data obtained from OLIS does not include the actual manufacturer name of the test, only that the test is a high-sensitivity assay (authors PK and DK oversaw this process). Accordingly, no sub-analyses were performed using specific assay cutoffs. However, some guidelines and key opinion leaders have advocated for a “low” troponin cutoff. For example, the 5 ng/L cutoff for the Abbott and Siemens hs-cTnI assays have been suggested as a universal cutoff to rule-out. Applying one such common low cutoff, as was done in this paper, yields sensitivity aspects that are inferior for hs-cTnI vs hs-cTnT (where there is only one manufacturer). We believe our approach re-emphasizes the importance of: i) using assay specific cutoffs; or ii) using the clinical chemistry score which has shown to yield higher diagnostic estimates regardless of the assay. Second, sex-specific cutoffs for hs-cTn were not used for rule-out. Sex-specific cutoffs have been advocated to detect myocardial injury (see on-going CODE-MI trial in Canada: https://clinicaltrials.gov/ct2/show/NCT03819894) and not to rule-out MI. There are minor differences in hs-cTn concentrations between the sexes within the low normal range, which are actually within the imprecision of the assay at the low end. As laboratories do not regularly monitor the low end and laboratory recommendations suggest testing only one quality control level in the normal range, analytically only one cutoff was used. Third, the largest group (n=880,518) excluded from our analyses was those that did not have all three laboratory tests reported. It is unclear if there are differences in risk and test performance between those with only hs-cTn tested versus those with two or three laboratory tests reported. Accordingly, the performance of Approach 1 in those patients that had only hs-cTn tested is unknown and caution is warranted in extrapolating findings from this study assessing all three laboratory tests to this group. Fourth, the percentage of patients deemed low-risk is smaller using Approach 2 versus Approach 1. This difference may result in more patients using Approach 2 being subjected to a second blood draw and a second hs-cTn measurement which is useful to document acute myocardial injury.
      Overall, the clinical chemistry score (Approach 2) yields higher sensitivity and represents a standardized, one blood draw and done approach, for classifying very lowrisk ED patients whose incidence of MI/death <1% over 1 year.
      Conflict of Interest Disclosures: Dr. Kavsak has received grants/reagents/consultant/advisor/ honoraria from Abbott Laboratories , Abbott Point of Care, Beckman Coulter, Ortho Clinical Diagnostics , Quidel, Randox Laboratories, Roche Diagnostics , Siemens Healthcare Diagnostics, and Thermo Fisher Scientific . McMaster University has filed patents with Drs. Kavsak and Worster listed as inventors in the acute cardiovascular biomarker field, in particular, a patent has been granted by the EPO related to the data presented in this study “METHOD OF DETERMINING RISK OF AN ADVERSE CARDIAC EVENT”. No other disclosures were reported.

      Uncited reference

      • Neumann J.T.
      • Twerenbold R.
      • Ojeda F.
      • et al.
      Application of High-Sensitivity Troponin in Suspected Myocardial Infarction.
      .

      Acknowledgements

      This study was supported by ICES, which is funded by an annual grant from the Ontario Ministry of Health (MOH) and the Ministry of Long-Term Care (MLTC). Content enclosed are based on data provided by CIHI. Analyses, opinions, and statements expressed herein are those of the authors and not necessarily those of CIHI; no endorsement is intended or should be inferred. We thank IQVIA Solutions Canada Inc. for use of their Drug Information File.

      References

      1. Gulati M, Levy PD, Mukherjee D, et al. 2021 AHA/ACC/ASE/CHEST/SAEM/SCCT/SCMR Guideline for the Evaluation and Diagnosis of Chest Pain: Executive Summary: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation 2021;144:e368-e454.c

        • Neumann J.T.
        • Twerenbold R.
        • Ojeda F.
        • et al.
        Application of High-Sensitivity Troponin in Suspected Myocardial Infarction.
        N Engl J Med. 2019; 380: 2529-2540
        • Lau G.
        • Koh M.
        • Kavsak P.A.
        • et al.
        Clinical outcomes for chest pain patients discharged home from emergency departments using high-sensitivity versus conventional cardiac troponin assays.
        Am Heart J. 2020; 221: 84-94
        • Kavsak P.A.
        • Cerasuolo J.O.
        • Ko D.T.
        • et al.
        Using the clinical chemistry score in the emergency department to detect adverse cardiac events: a diagnostic accuracy study.
        CMAJ Open. 2020; 8: E676-E684
        • Westwood M.E.
        • Armstrong N.
        • Worthy G.
        • et al.
        Optimizing the Use of High-Sensitivity Troponin Assays for the Early Rule-out of Myocardial Infarction in Patients Presenting with Chest Pain: A Systematic Review.
        Clin Chem. 2021; 67: 237-244