Author + information
- Peter P. Swoboda, MBBS,
- Adam K. McDiarmid, MBBS,
- Bara Erhayiem, BMBS,
- David A. Broadbent, MSc,
- Laura E. Dobson, MBChB,
- Pankaj Garg, MD,
- Carrie Ferguson, PhD,
- Stephen P. Page, MD,
- John P. Greenwood, PhD and
- Sven Plein, PhD∗ ()
- ↵∗Multidisciplinary Cardiovascular Research Centre and Division of Biomedical Imaging, Leeds Institute of Cardiovascular and Metabolic Medicine, University of Leeds, Clarendon Way, Leeds, LS2 9JT, United Kingdom
Athletes who train regularly can develop left ventricular (LV) hypertrophy, which can be difficult to differentiate from hypertrophic cardiomyopathy (HCM), the leading cause of sudden cardiac death in young athletes. Current guidelines advise that patients with HCM should avoid competitive sport, and it is therefore imperative that HCM and physiological remodeling are correctly identified. Cardiovascular magnetic resonance (CMR) T1 and extracellular volume (ECV) mapping provide quantitative assessment of myocardial composition. We hypothesized that ECV could differentiate athletic from pathological hypertrophy, particularly in subjects with indeterminate maximum wall thickness, defined as 12 to 15 mm (1).
Fifty HCM patients, 40 athletes and 35 sedentary volunteers, underwent 3.0-T CMR including 5 beats (3 seconds) 3 beats Modified Look-Locker Inversion (MOLLI) T1 maps before and 15 min after administration of 0.15 mmol/kg intravenous gadobutrol. HCM was diagnosed independently according to current guidelines (1). The 40 competitive athletes (11 runners, 13 triathletes, and 16 cyclists) trained >6 h per week, had mean maximal oxygen consumption (Vo2 max) concentration of 58.3 ± 9.0 ml/min/kg and were <45 years of age. Sedentary volunteers exercised <3 h per week. The study was approved by the local ethics committee (14/YH/0126).
Analysis was carried out using cvi42 software (Circle CVI, Calgary, Canada). Maximum wall thickness was measured from diastolic short-axis cine images, and native T1 and ECV were measured in the thickest segment. ECV was calculated from hematocrit, native, and post-contrast T1 times of myocardium and blood pool (2). A Mann-Whitney U test was used to compare athletes and HCM subjects. Receiver operating characteristic analysis was used to determine the diagnostic accuracies (SPSS version 20.00 software; IBM Corp., Armonk, New York).
Both native T1 and ECV of the thickest segment were lower in athletes than HCM (1182.7 ± 42.4 ms vs. 1261.0 ± 66.0 ms and 22.7 ± 3.3% vs. 32.3 ± 7.9% of subjects, respectively; p < 0.001 for both). Two athletes (5%) had subepicardial lateral late gadolinium enhancement (LGE) in a myocarditis pattern, no controls had LGE, and 35 HCM subjects (70%) had LGE.
ECV of the thickest segment was significantly lower in athletes than in controls (22.7 ± 3.3% vs. 24.3 ± 2.6%, respectively; p = 0.006). Differences between native T1 in athletes and that in controls did not reach statistical significance (p = 0.18).
In athletes, there were significant negative correlations between ECV and maximum segment thickness (R = −0.40; p = 0.01) and LV mass (R = −0.37; p = 0.02) (Figure 1). In controls there were also significant negative correlations between ECV and maximum segment thickness (R = −0.45; p < 0.01), and LV mass (R = −0.42; p = 0.01). In HCM subjects there was a significant positive correlation between ECV and maximum segment thickness (Spearman's rank [Rs] = 0.43; p = 0.002) but not LV mass (p = 0.33). For athletes, controls, and HCM subjects there were no significant correlations among native T1 and maximum segment thickness or LV mass.
To detect the 50 cases of HCM from the 40 athletes, the diagnostic accuracy (area under the curve [AUC]) of maximal segment thickness, native T1, and ECV were 0.986 (95% confidence interval [CI]: 0.935 to 0.999), 0.847 (95% CI: 0.756 to 0.914), and 0.936 (95% CI: 0.864 to 0.977), respectively (p < 0.001 for all). There were no significant differences between AUCs. The AUC of LGE to diagnose HCM correctly was 0.825 (range: 0.731 to 0.897; p <0.001; sensitivity: 70%; specificity: 83%). The AUC of ECV was superior to that of LGE (p = 0.004), although the differences between native T1 and LGE were nonsignificant (p = 0.66).
In 26 subjects (10 athlete and 16 HCM subjects), the maximum segment thickness fell in the intermediate range of 12 to 15 mm. In these subjects, native T1 in the thickest segment was 1170.6 ± 34.8 ms in athletes versus 1251.9 ± 47.2 ms in HCM subjects (p < 0.001) and ECV was 21.1 ± 3.2% in athletes versus 28.5 ± 2.3% in HCM subjects (p <0.001). AUCs to correctly detect HCM for native T1 and ECV were 0.938 (95% CI: 0.769 to 0.995) and 0.963 (95% CI: 0.805 to 0.999; p <0.001 for both). There were no significant differences between AUCs. The optimal cutoff values used to diagnose HCM were ECV >22.5% (sensitivity: 100%; specificity: 90%) and native T1 >1,217.6 ms (sensitivity: 81%; specificity: 100%).
As LV hypertrophy increases there is a reduction in ECV in athletes but an increase in ECV in patients with HCM. Based on this divergent finding, ECV can be used to distinguish HCM and athletic remodelling with high diagnostic accuracy, particularly in subjects with indeterminate maximal wall thickness. The negative correlation between ECV and wall thickness in athletes and sedentary controls suggests that the increase in LV mass in healthy myocardium is mediated by cellular hypertrophy, whereas in HCM, it is mediated by cellular disarray and extracellular matrix expansion. CMR using T1 mapping, thus, has a potential role in the exclusion of HCM in athletes presenting with left ventricular hypertrophy. These findings need further study in more varied populations of athletes and patients with HCM who partake in competitive sport.
Please note: Drs. Swoboda and Plein are funded by British Heart Foundation fellowships (FS/12/88/29901 and FS/1062/28409, respectively). This study was supported by the National Institute for Health Research (NIHR), Leeds Clinical Research Facility. The views expressed are those of the authors and not necessarily those of the National Health Service, NIHR, or the Department of Health. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Drs. Swoboda and McDiarmid contributed equally to this work.
- American College of Cardiology Foundation
- Gersh B.J.,
- Maron B.J.,
- Bonow R.O.,
- et al.
- Flett A.S.,
- Hayward M.P.,
- Ashworth M.T.,
- et al.