Author + information
- Published online October 2, 2017.
- Amardeep Ghosh Dastidar, MBBS(Hons),
- Jonathan C.L. Rodrigues, BSc(Hons), MBChB(Hons),
- Thomas W. Johnson, BSc(Hons), MBBS, MD,
- Estefania De Garate, MBChB,
- Priyanka Singhal, MBChB,
- Anna Baritussio, MD,
- Alessandra Scatteia, MD,
- Julian Strange, MBChB, MD,
- Angus K. Nightingale, MA, MB BChir, MD,
- Gianni D. Angelini, MD MCh,
- Andreas Baumbach, MD,
- Victoria Delgado, MD, PhD and
- Chiara Bucciarelli-Ducci, MD, PhD∗ ()
- ↵∗NIHR Bristol Cardiovascular Biomedical Research Unit, CMR Unit, Bristol Heart Institute, Upper Maudlin Street, Bristol, BS2 8HW, United Kingdom
Seven to 15% of patients with acute coronary syndrome (ACS) have nonobstructed coronary arteries, an entity that is known as myocardial infarction with nonobstructed coronary arteries (MINOCA) (1). In these patients, cardiac magnetic resonance (CMR) can identify different underlying etiologies (2). However, the optimum timing and the impact of CMR on clinical management are unknown. We aimed to evaluate the diagnostic and decision-making implications of CMR timing (“early” ≤2 weeks vs. “late” >2 weeks after presentation) in MINOCA.
A total of 204 consecutive patients (56 ± 17 years; 51% men) with troponin-positive ACS (as per the European Society of Cardiology guidelines for ST-segment elevation [STE] or non-STE ACS and the Third Universal definition of myocardial infarction [MI]) and unobstructed coronary arteries (MINOCA) with unclear final diagnoses were referred for CMR and included in the study from September 2011 to July 2014. Nineteen percent presented with ST-segment elevation on the electrocardiogram (ECG) and the remaining patients presented with non–ST-segment elevation myocardial infarction. The mean troponin level was 640 ng/l (normal <14 ng/l). The study was reviewed and approved by local institutional review board. CMR (1.5-T) was performed using a comprehensive protocol (cines, T2-weighted, and late gadolinium enhancement [LGE] sequences). Myocarditis was diagnosed using the Lake Louise Criteria, MI was diagnosed by territorial subendocardial and/or transmural LGE, and Takotsubo cardiomyopathy was diagnosed by modified Mayo Clinic criteria.
Pre-CMR diagnosis was defined per the referring clinician’s suspected diagnosis recorded in the CMR referral, based on a composite of clinical, biomarkers, ECG, and echocardiographic and angiographic information, which reflects current clinical practice and guideline recommendations.
Post-CMR diagnosis was also recorded and was subsequently compared with the pre-CMR diagnosis to investigate whether the results confirmed the pre-CMR diagnosis or identified a new diagnosis.
Significant clinical impact was defined as identification of a new diagnosis or a change in management. For changes in management, we recorded: 1) changes in length of hospital stay (shorter and/or longer); 2) changes in discharge medications (introduction and/or discontinuation); or 3) introduction and/or avoidance of additional invasive procedures.
To identify the correlates of clinical and CMR indexes with a significant clinical impact, we performed univariate and multivariate regression analysis for age, sex, troponin T, indexed left ventricular (LV) end-diastolic volume, LV ejection fraction, regional wall motion abnormality, myocardial edema, and LGE. Propensity matching, using a nonparsimonious multivariable logistic regression model, was performed to minimize any selection bias between early and late CMR groups. SPSS version 23 (IBM, Armonk, New York) was used for statistical analysis.
Overall, CMR provided a final diagnosis in 70% (n = 143) of patients (myocarditis: 27%; MI: 26%; Takotsubo cardiomyopathy: 9%; and other cardiomyopathies: 9%). In 30% (n = 61) of patients, CMR demonstrated a structurally normal heart (no RWMA, edema, or scarring). The diagnostic yield of CMR was significantly higher when performed early (84% vs. 57%; p < 0.0001). Myocarditis (33%) was the most common diagnosis in the early group, whereas MI was the most common diagnosis in the late group (26%). The detection of myocarditis (33% vs. 21%; p = 0.04) and Takotsubo cardiomyopathy (16% vs. 3%; p = 0.002) was also significantly higher in patients who underwent early CMR, whereas no change was observed for MI (26% vs. 26%; p = 1).
Overall, CMR had a significant clinical impact in 66% (n = 134) of patients (new diagnosis: 54%; change in management: 41%) (Figure 1). CMR led to subsequent invasive procedures in 5% (myocardial biopsy: n = 5; implantable cardiac defibrillator: n = 3; ventricular assist device: n = 2); 4% of patients avoided an invasive procedure. Age (odds ratio [OR]: 1.024; 95% confidence interval [CI]: 1.006 to 1.041; p = 0.008), myocardial edema (OR: 1.765; 95% CI: 0.938 to 3.323; p = 0.078), and LGE (OR: 2.393; 95% CI: 1.318 to 4.345; p = 0.004) were significant univariate predictors of clinical impact (p < 0.1, considered significant for univariate analysis). In a multivariate model, only age (OR: 1.035; 95% CI: 1.013 to 1.058; p = 0.002) and LGE (OR: 2.411; 95% CI: 1.17 to 4.968; p = 0.017) remained significant.
Propensity score matching identified 58 pairs of early and late subjects. The results confirmed the significantly higher diagnostic yield in the propensity-matched early group versus late group (88% vs. 50%; p < 0.0001). The clinical impact also improved significantly in the propensity-matched early CMR group (76% vs. 51%; p = 0.01).
The study demonstrated the importance of performing CMR early in MINOCA, which provides a window of opportunity to image myocardial damage before healing occurs, thereby maximizing the diagnostic yield. This is particularly relevant in potentially reversible conditions (acute myocarditis and Takotsubo cardiomyopathy).
The timing of CMR was partly based on the referring physician’s discretion or scanner availability. However, with propensity scoring, we matched the early CMR group with the late CMR group, thereby reducing the selection bias. CMR-guided diagnosis was assumed to be correct based on previous literature (2).
In conclusion, in consecutive patients with MINOCA, CMR established a definitive diagnosis in 70% of patients and made a significant additive impact on diagnosis and/or clinical management in 66% of patients, with LGE being the best independent predictor of clinical impact more than the traditional clinical and diagnostic markers. Moreover, the diagnostic value and the clinical impact of CMR were highest when performed within 2 weeks from presentation.
Please note: This study was funded by the NIHR Biomedical Research Unit in Cardiovascular Disease at the University Hospitals Bristol NHS Foundation Trust and the University of Bristol. The views expressed are those of the authors and not necessarily those of the UK National Health Service, National Institute for Health Research, or Department of Health. Dr. Bucciarelli-Ducci is a consultant for Circle Cardiovascular Imaging. Dr. Delgado has received speaker fees from Abbott Vascular. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- 2017 American College of Cardiology Foundation