Exercise cardiac magnetic resonance to differentiate athlete’s heart from structural heart disease

Exercise cardiac magnetic resonance to differentiate athlete’s heart from structural heart disease Abstract Aims The distinction between left ventricular (LV) dilation with mildly reduced LV ejection fraction (EF) in response to regular endurance exercise training and an early cardiomyopathy is a frequently encountered and difficult clinical conundrum. We hypothesized that exercise rather than resting measures would provide better discrimination between physiological and pathological LV remodelling and that preserved exercise capacity does not exclude significant LV damage. Methods and results We prospectively included 19 subjects with LVEF between 40 and 52%, comprising 10 ostensibly healthy endurance athletes (EA-healthy) and nine patients with dilated cardiomyopathy (DCM). In addition, we recruited five EAs with a region of subepicardial LV. Receiver operating characteristic fibrosis (EA-fibrosis). Cardiac magnetic resonance (CMR) imaging was performed at rest and during supine bicycle exercise. Invasive afterload measures were obtained to enable calculations of biventricular function relative to load (an estimate of contractility). In DCM and EA-fibrosis subjects there was diminished augmentation of LVEF (5 ± 6% vs. 4 ± 3% vs. 14 ± 3%; P = 0.001) and contractility [LV end-systolic pressure–volume ratio, LVESPVR; 1.4 (1.3–1.6) vs. 1.5 (1.3–1.6) vs. 1.8 (1.7–2.7); P < 0.001] during exercise relative to EA-healthy. Receiver-operator characteristic curves demonstrated that a cut-off value of 11.2% for ΔLVEF differentiated DCM and EA-fibrosis patients from EA-healthy [area under the curve (AUC) = 0.92, P < 0.001], whereas resting LVEF and VO2max were not predictive. The AUC value for LVESPVR ratio was similar to that of ΔLVEF. Conclusions Functional cardiac evaluation during exercise is a promising tool in differentiating healthy athletes with borderline LVEF from those with an underlying cardiomyopathy. Excellent exercise capacity does not exclude significant LV damage. athlete’s heart , dilated cardiomyopathy , exercise , myocardial fibrosis , cardiac magnetic resonance imaging , contractile reserve Introduction Profound left ventricular (LV) dilation with mildly reduced LV ejection fraction (EF) at rest is seen in >10% of elite endurance athletes (EA).1 Distinguishing this phenotype from true underlying LV systolic dysfunction can be a difficult clinical conundrum1 with important consequences for competitive sport participation.2 The two conditions are sometimes considered mutually exclusive based on the assumption that excellent exercise capacity precludes significant pathology. However, we and others have described athletes with significant pathology diagnosed whilst continuing to compete at the highest level.3–6 Thus, new methods are required to accurately separate health from pathology amongst athletes with ambiguous cardiac function at rest. Guidelines suggest that exercise assessments of cardiac function can be used to differentiate athlete’s heart from cardiomyopathy when resting measures of function appear abnormal.7 Although a logical premise, there have not been any studies to support this assertion and the accuracy of exercise imaging is uncertain. We have previously demonstrated that real-time cardiac magnetic resonance imaging performed during exercise (ex-CMR) can be used to accurately quantify cardiac function and represents the ideal modality for investigating differences in contractile reserve between normal and diseased myocardium.8–10 This study sought to evaluate the cardiac response to dynamic exercise in ostensibly healthy EA with mildly reduced LVEF as compared with a group of non-athletic subjects with mild or early dilated cardiomyopathy (DCM). Furthermore, we tested whether ex-CMR was useful to identify sub-clinical cardiac damage. Therefore, we included a third group of highly trained athletes in whom CMR had incidentally detected significant LV fibrosis while they were still competing at an elite level with normal exercise capacity. We hypothesized that LV contractile reserve would be able to discriminate between healthy athletes and apparently normal athletic subjects with underlying LV damage. Methods Subjects Thirteen EA (all male) with low normal resting LVEF were recruited using a prospective inclusion criterion of LVEF ≤52% measured by CMR, in accordance with previous studies.1 These EA-healthy athletes were recruited from volunteers responding to advertisements at local triathlon and cycling clubs (n = 4) and from individuals referred to our institution (n = 9) after screening examinations revealed a mildly reduced LV systolic function in the absence of other signs of structural heart disease (i.e. normal wall thickness, no valvular heart disease, no regional wall motion abnormalities, and no signs of myocardial fibrosis). EA-healthy subjects were included if: (i) they were participating in regular cycling and/or running training of >6 h/week, (ii) they had no history of cardiovascular disease, and (iii) CMR at inclusion confirmed an LVEF ≤52%. Three athletes were excluded because CMR at enrolment revealed a normal LVEF prior to study participation. Therefore, the final EA-healthy cohort consisted of 10 subjects. Nine patients with mild DCM (eight male) and five EA with fibrosis (EA-fibrosis, all male) were recruited from an existing database in addition to new cases presenting over the study period. The DCM cohort consisted of seven first-degree family members of DCM patients without an identifiable mutation and two patients with partially recovered severe DCM. All subjects had LVEF ≥40 and ≤52% measured by CMR prior to study participation. The cohort of EA-fibrosis subjects consisted of high-level EA in whom significant delayed gadolinium enhancement (11.4 ± 4.7% of LV mass) was detected after screening evaluation revealed pathological T-wave inversion or ventricular arrhythmias, as previously described.4 Relative to our previous study cohort,4 only those EA-fibrosis subjects referred to our institution were included in this study, as well as two new cases presenting during the study period. Therefore, the final cohort consisted of five EA-fibrosis subjects. An example of an EA-fibrosis subject is shown in Figure 1. All of the EA-fibrosis subjects developed ventricular arrhythmias resulting in exclusion from competitive sports participation. Figure 1 View largeDownload slide Example of an endurance athlete with significant subepicardial delayed gadolinium enhancement (arrowheads). Figure 1 View largeDownload slide Example of an endurance athlete with significant subepicardial delayed gadolinium enhancement (arrowheads). To ascertain whether significant LV damage and elite exercise capacity could coexist, contractile reserve in the EA-fibrosis cohort was compared against a subset of EA-healthy subjects, matched for exercise capacity. The study protocol conformed to the Declaration of Helsinki and was approved by the Ethics Committee of UZ Leuven (B322201214035). All subjects provided informed consent. Study design Prior to CMR evaluation, a 20-gauge arterial catheter was placed in the radial artery for the measurement of systemic arterial pressures. Subsequently, biventricular volumes were measured during supine cycling exercise using a real-time CMR method that we previously described and validated against invasive standards.10 In brief, subjects performed supine exercise within the CMR bore using a cycle ergometer with adjustable electronic resistance (Lode, Groningen, The Netherlands). Images were acquired using a Philips Achieva 1.5 T CMR with a five-element phased-array coil (Philips Medical Systems, Best, The Netherlands) at rest and during supine bicycle exercise at 25%, 50%, and 66% of maximal power determined by previous upright cardiopulmonary exercise testing.10 Steady-state free precession cine imaging was performed without cardiac gating. Imaging parameters were: field of view 320 × 260 mm (approximately), 128 × 128 matrix, flip angle 50°, SENSE factor 2 (Cartesian k-space under-sampling), repetition time 1.8 ms, echo time 0.9 ms, and reconstructed voxel size 2.3 × 2.3 × 8 mm. A 3D stack of 13–18 contiguous 8 mm image slices, covering both ventricles from apex to base, was serially acquired in the short-axis plane and subsequently in the horizontal long-axis plane. All image frames were acquired during free breathing with a temporal resolution of 36–38 ms. Systemic arterial pressure measurements were continuously recorded during the exercise CMR protocol and analysed off-line using LabChart v6.1.1 (AD Instruments). Using in-house developed software (RightVol, Leuven, Belgium), LV and right ventricular (RV) end-diastolic volumes and end-systolic volumes (EDVi, ESVi) and left atrial and right atrial (LA, RA) maximal volumes and minimal volumes (Vmax, Vmin) were calculated by a summation of disks and indexed for body surface area. LVEF and RVEF were calculated as (EDVi–ESVi)/EDVi. Stroke volume was measured as EDVi – ESVi and cardiac index (CI) as the product of SVi and heart rate. If assuming that V0 (zero-volume intercept of the end-systolic pressure–volume relationship) is negligible, a single point LV end-systolic pressure–volume ratio (LVESPVR) can be calculated using the formula (0.9*systolic blood pressure)/LVESV as a surrogate of ventricular elastance.9,11 LV contractile reserve was defined as a ratio of peak-exercise to resting LVESPVR (subsequently referred to as ‘LVESPVR ratio’). Arterial elastance (Ea) was calculated as (0.9*systolic blood pressure)/LVSV.11 As a measure of global atrial function, atrial total emptying fraction (LAEF, RAEF) was calculated as [(Vmax – Vmin)/Vmax].12,13 NT-proBNP was analysed from venous blood samples. Statistical analysis Data were analysed using IBM SPSS statistics 22 software. Gaussian distribution of all continuous variables was tested using a Kolmogorov–Smirnov test. Descriptive data for continuous variables are presented as mean ± standard deviation or as medians (25% and 75% percentile) as appropriate. Comparisons between groups for continuous variables were performed by one-way analysis of variance (ANOVA) or the Kruskal–Wallis with Bonferroni or Dunn’s test for multiple comparison posthoc correction, as appropriate. The Fisher’s exact or the χ2 test was used for categorical variables. To determine the sample sizes, the following estimates were used: in a previous study using exercise CMR, we demonstrated that healthy EA had an 8 ± 6% increase in LVEF from rest to maximal exercise.9 According to our hypothesis, we predicted that LVEF will not change (0% increase) during exercise in the subjects with underlying LV damage.14,15 Using these assumptions, a sample size of n = 9 was calculated to provide 80% power in detecting impaired LVEF augmentation during exercise in the groups with LV pathology (α = 5%, 1−β = 80%, n = 9). The biventricular volume response from rest to peak-intensity exercise in the different groups was compared using repeated measures ANOVA with exercise-intensity as within-subject effect and group (DCM vs. EA-fibrosis vs. EA-healthy) as a between-subject effect. Receiver operating characteristic (ROC) curves were constructed to determine the diagnostic accuracy of resting and exercise measures of LV function16 for distinguishing EA-healthy subjects from both non-athletic and athletic subjects with underlying LV damage (expressed as area and 95% confidence intervals). The ‘optimal’ cut-off value for each parameter was defined as the value of the parameter that corresponded with the highest sum of specificity and sensitivity. The significance of differences in area under the curve (AUC) of the correlated rest and peak exercise ROC curves was tested using the methodology described by DeLong et al.17 A P-value <0.05 was considered statistically significant. Results The demographic, clinical characteristics, and cardiopulmonary exercise testing data are presented in Table 1. All groups were of similar age and gender. As expected, both EA-healthy and EA-fibrosis cohorts had superior exercise capacity compared with DCM patients. A majority of DCM patients received therapy with a beta-blocker and angiotensin converting enzyme (ACE) inhibitor. All negative chronotropic medication was withheld for at least 24 h prior to exercise testing. NT-proBNP tended to be higher in DCM patients compared with the other groups. Invasive pressure measurement was performed in 20 of 24 study participants (6/10 EA-healthy, 9/9 DCM, and 5/5 EA-fibrosis). All EA-healthy subjects and DCM patients had resting LVEF <55% at the time of the exercise CMR protocol (noting some variability from the measure used for study inclusion). All of the EA-fibrosis subjects had significant delayed gadolinium enhancement (11.4 ± 4.7% of LV mass) as compared with two DCM subjects and none of the EA-healthy subjects. One of the EA-fibrosis subjects also had delayed gadolinium enhancement (DGE) of the RV free wall. Table 1 Baseline characteristics EA-healthy (n = 10) DCM (n = 9) EA-fibrosis (n = 5) ANOVA P-value Clinical  Age (years) 34 ± 12 44 ± 14 33 ± 8 0.149  BMI (kg/m2) 23.8 ± 2.3 25.1 ± 3.8 22.0 ± 0.7 0.173  Weight (kg) 77.9 ± 9.2 78.2 ± 16.2 78.1 ± 6.0 0.998  Male, n 9 8 5 1.00 Medication  Beta-blockers, n 0 6 1 0.005  ACE inhibitors, n 0 6 0 0.001  AR blockers, n 0 2 0 0.162 Biochemical  NT-proBNP (pg/mL) 32 (14–49) 146 (51–320) 57 (19–83) 0.062 CPET  VO2peak (mL/min) 4343 ± 760 2253 ± 534* 5343 ± 532*,† <0.001  VO2peak (mL/min/kg) 56.2 ± 10.1 29.0 ± 5.3* 68.4 ± 4.2*,† <0.001  Peak power (watts) 389 ± 70 206 ± 55* 444 ± 56† <0.001  Peak HR (b.p.m.) 182 ± 8 164 ± 19* 176 ± 6 0.031 CMR at inclusion  LVEF (%) 49.7 ± 1.9 47.7 ± 5.5 53.0 ± 9.2 0.243  DGE, n 0 2 5 <0.001 EA-healthy (n = 10) DCM (n = 9) EA-fibrosis (n = 5) ANOVA P-value Clinical  Age (years) 34 ± 12 44 ± 14 33 ± 8 0.149  BMI (kg/m2) 23.8 ± 2.3 25.1 ± 3.8 22.0 ± 0.7 0.173  Weight (kg) 77.9 ± 9.2 78.2 ± 16.2 78.1 ± 6.0 0.998  Male, n 9 8 5 1.00 Medication  Beta-blockers, n 0 6 1 0.005  ACE inhibitors, n 0 6 0 0.001  AR blockers, n 0 2 0 0.162 Biochemical  NT-proBNP (pg/mL) 32 (14–49) 146 (51–320) 57 (19–83) 0.062 CPET  VO2peak (mL/min) 4343 ± 760 2253 ± 534* 5343 ± 532*,† <0.001  VO2peak (mL/min/kg) 56.2 ± 10.1 29.0 ± 5.3* 68.4 ± 4.2*,† <0.001  Peak power (watts) 389 ± 70 206 ± 55* 444 ± 56† <0.001  Peak HR (b.p.m.) 182 ± 8 164 ± 19* 176 ± 6 0.031 CMR at inclusion  LVEF (%) 49.7 ± 1.9 47.7 ± 5.5 53.0 ± 9.2 0.243  DGE, n 0 2 5 <0.001 BMI, body mass index, AR, angiotensin II receptor; CPET, cardiopulmonary exercise testing; DGE, delayed gadolinium enhancement. * P < 0.05 for difference vs. EA-healthy. †P < 0.05 for difference vs. DCM. Table 1 Baseline characteristics EA-healthy (n = 10) DCM (n = 9) EA-fibrosis (n = 5) ANOVA P-value Clinical  Age (years) 34 ± 12 44 ± 14 33 ± 8 0.149  BMI (kg/m2) 23.8 ± 2.3 25.1 ± 3.8 22.0 ± 0.7 0.173  Weight (kg) 77.9 ± 9.2 78.2 ± 16.2 78.1 ± 6.0 0.998  Male, n 9 8 5 1.00 Medication  Beta-blockers, n 0 6 1 0.005  ACE inhibitors, n 0 6 0 0.001  AR blockers, n 0 2 0 0.162 Biochemical  NT-proBNP (pg/mL) 32 (14–49) 146 (51–320) 57 (19–83) 0.062 CPET  VO2peak (mL/min) 4343 ± 760 2253 ± 534* 5343 ± 532*,† <0.001  VO2peak (mL/min/kg) 56.2 ± 10.1 29.0 ± 5.3* 68.4 ± 4.2*,† <0.001  Peak power (watts) 389 ± 70 206 ± 55* 444 ± 56† <0.001  Peak HR (b.p.m.) 182 ± 8 164 ± 19* 176 ± 6 0.031 CMR at inclusion  LVEF (%) 49.7 ± 1.9 47.7 ± 5.5 53.0 ± 9.2 0.243  DGE, n 0 2 5 <0.001 EA-healthy (n = 10) DCM (n = 9) EA-fibrosis (n = 5) ANOVA P-value Clinical  Age (years) 34 ± 12 44 ± 14 33 ± 8 0.149  BMI (kg/m2) 23.8 ± 2.3 25.1 ± 3.8 22.0 ± 0.7 0.173  Weight (kg) 77.9 ± 9.2 78.2 ± 16.2 78.1 ± 6.0 0.998  Male, n 9 8 5 1.00 Medication  Beta-blockers, n 0 6 1 0.005  ACE inhibitors, n 0 6 0 0.001  AR blockers, n 0 2 0 0.162 Biochemical  NT-proBNP (pg/mL) 32 (14–49) 146 (51–320) 57 (19–83) 0.062 CPET  VO2peak (mL/min) 4343 ± 760 2253 ± 534* 5343 ± 532*,† <0.001  VO2peak (mL/min/kg) 56.2 ± 10.1 29.0 ± 5.3* 68.4 ± 4.2*,† <0.001  Peak power (watts) 389 ± 70 206 ± 55* 444 ± 56† <0.001  Peak HR (b.p.m.) 182 ± 8 164 ± 19* 176 ± 6 0.031 CMR at inclusion  LVEF (%) 49.7 ± 1.9 47.7 ± 5.5 53.0 ± 9.2 0.243  DGE, n 0 2 5 <0.001 BMI, body mass index, AR, angiotensin II receptor; CPET, cardiopulmonary exercise testing; DGE, delayed gadolinium enhancement. * P < 0.05 for difference vs. EA-healthy. †P < 0.05 for difference vs. DCM. Cardiac response to exercise Resting and peak exercise cardiac haemodynamics are shown in Table 2. LV and RV end-diastolic volumes and RV end-systolic volumes were smaller in DCM compared with the EA-healthy and EA-fibrosis, whereas LV end-systolic volumes were similar. Table 2 Cardiac haemodynamics EA-healthy (n = 10) DCM (n = 9) EA-fibrosis (n = 5) P-value Heart rate (b.p.m.) Rest 66 ± 11 67 ± 8 52 ± 9*,† 0.025 Peak ex 153 ± 20 142 ± 13 138 ± 20 0.238 Mean arterial pressure (mmHg) Rest 88 ± 7 90 ± 10 81 ± 11 0.317 Peak ex 113 ± 14 117 ± 15 109 ± 7 0.580 Ea (mmHg/mL) Rest 1.9 ± 0.4 2.6 ± 0.6* 1.6 ± 0.4 0.003 Peak ex 2.1 ± 0.5 2.9 ± 0.3* 2.1 ± 0.3 <0.001 LVEDV (mL/m2) Rest 122 ± 18 99 ± 23* 133 ± 11† 0.007 Peak ex 124 ± 23 104 ± 26 134 ± 15 0.058 RVEDV (mL/m2) Rest 124 ± 22 81 ± 18* 125 ± 19† <0.001 Peak ex 119 ± 23 77 ± 20* 122 ± 22† <0.001 LVESV (mL/m2) Rest 59 ± 9† 52 ± 15 63 ± 17 0.272 Peak ex 44 ± 9 51 ± 22 58 ± 17 0.298 RVESV (mL/m2) Rest 59 ± 10 34 ± 10* 58 ± 2† <0.001 Peak ex 40 ± 9 26 ± 10 51 ± 17† 0.003 LVSV (mL/m2) Rest 63 ± 11 47 ± 10* 70 ± 9† 0.001 Peak ex 81 ± 15 53 ± 9* 76 ± 8† <0.001 RVSV (mL/m2) Rest 64 ± 11 47 ± 9* 67 ± 8† 0.001 Peak ex 79 ± 15 51 ± 10* 71 ± 7† <0.001 LVEF (%) Rest 51.1 ± 2.8 47.7 ± 5.5 53.0 ± 9.2 0.215 Peak ex 64.8 ± 3.7 52.8 ± 9.5* 57.2 ± 8.9 0.007 RVEF (%) Rest 51.9 ± 2.1 58.4 ± 6.6* 53.9 ± 3.3 0.017 Peak ex 66.3 ± 4.0 66.6 ± 5.9 59.0 ± 7.5 0.047 CI (L/min/m2) Rest 4.3 ± 1.4 3.1 ± 0.6 3.5 ± 0.5 0.070 Peak ex 12.3 ± 3.4 7.4 ± 1.6* 10.1 ± 1.1 0.001 LAVmax (mL/m2) Rest 52 ± 8 46 ± 15 57 ± 11 0.219 Peak ex 56 ± 18 53 ± 15 61 ± 20 0.745 RAVmax (mL/m2) Rest 78 ± 9 63 ± 26 95 ± 18† 0.020 Peak ex 68 ± 22 66 ± 29 76 ± 22 0.755 LAVmin (mL/m2) Rest 27 ± 5 24 ± 8 36 ± 10† 0.020 Peak ex 25 ± 8 31 ± 13 36 ± 11 0.202 RAVmin (mL/m2) Rest 42 ± 8 34 ± 15 59 ± 10*,† 0.003 Peak ex 26 ± 9 34 ± 18 40 ± 13 0.188 EA-healthy (n = 10) DCM (n = 9) EA-fibrosis (n = 5) P-value Heart rate (b.p.m.) Rest 66 ± 11 67 ± 8 52 ± 9*,† 0.025 Peak ex 153 ± 20 142 ± 13 138 ± 20 0.238 Mean arterial pressure (mmHg) Rest 88 ± 7 90 ± 10 81 ± 11 0.317 Peak ex 113 ± 14 117 ± 15 109 ± 7 0.580 Ea (mmHg/mL) Rest 1.9 ± 0.4 2.6 ± 0.6* 1.6 ± 0.4 0.003 Peak ex 2.1 ± 0.5 2.9 ± 0.3* 2.1 ± 0.3 <0.001 LVEDV (mL/m2) Rest 122 ± 18 99 ± 23* 133 ± 11† 0.007 Peak ex 124 ± 23 104 ± 26 134 ± 15 0.058 RVEDV (mL/m2) Rest 124 ± 22 81 ± 18* 125 ± 19† <0.001 Peak ex 119 ± 23 77 ± 20* 122 ± 22† <0.001 LVESV (mL/m2) Rest 59 ± 9† 52 ± 15 63 ± 17 0.272 Peak ex 44 ± 9 51 ± 22 58 ± 17 0.298 RVESV (mL/m2) Rest 59 ± 10 34 ± 10* 58 ± 2† <0.001 Peak ex 40 ± 9 26 ± 10 51 ± 17† 0.003 LVSV (mL/m2) Rest 63 ± 11 47 ± 10* 70 ± 9† 0.001 Peak ex 81 ± 15 53 ± 9* 76 ± 8† <0.001 RVSV (mL/m2) Rest 64 ± 11 47 ± 9* 67 ± 8† 0.001 Peak ex 79 ± 15 51 ± 10* 71 ± 7† <0.001 LVEF (%) Rest 51.1 ± 2.8 47.7 ± 5.5 53.0 ± 9.2 0.215 Peak ex 64.8 ± 3.7 52.8 ± 9.5* 57.2 ± 8.9 0.007 RVEF (%) Rest 51.9 ± 2.1 58.4 ± 6.6* 53.9 ± 3.3 0.017 Peak ex 66.3 ± 4.0 66.6 ± 5.9 59.0 ± 7.5 0.047 CI (L/min/m2) Rest 4.3 ± 1.4 3.1 ± 0.6 3.5 ± 0.5 0.070 Peak ex 12.3 ± 3.4 7.4 ± 1.6* 10.1 ± 1.1 0.001 LAVmax (mL/m2) Rest 52 ± 8 46 ± 15 57 ± 11 0.219 Peak ex 56 ± 18 53 ± 15 61 ± 20 0.745 RAVmax (mL/m2) Rest 78 ± 9 63 ± 26 95 ± 18† 0.020 Peak ex 68 ± 22 66 ± 29 76 ± 22 0.755 LAVmin (mL/m2) Rest 27 ± 5 24 ± 8 36 ± 10† 0.020 Peak ex 25 ± 8 31 ± 13 36 ± 11 0.202 RAVmin (mL/m2) Rest 42 ± 8 34 ± 15 59 ± 10*,† 0.003 Peak ex 26 ± 9 34 ± 18 40 ± 13 0.188 LAVmax, left atrial maximal volume; LAVmin, left atrial minimal volume; RAVmax, right atrial maximal volume; RAVmin, right atrial minimal volume; RVEDV, right ventricular end-diastolic volume. * P < 0.05 for difference vs. EA-healthy. †P < 0.05 for difference vs. DCM. Table 2 Cardiac haemodynamics EA-healthy (n = 10) DCM (n = 9) EA-fibrosis (n = 5) P-value Heart rate (b.p.m.) Rest 66 ± 11 67 ± 8 52 ± 9*,† 0.025 Peak ex 153 ± 20 142 ± 13 138 ± 20 0.238 Mean arterial pressure (mmHg) Rest 88 ± 7 90 ± 10 81 ± 11 0.317 Peak ex 113 ± 14 117 ± 15 109 ± 7 0.580 Ea (mmHg/mL) Rest 1.9 ± 0.4 2.6 ± 0.6* 1.6 ± 0.4 0.003 Peak ex 2.1 ± 0.5 2.9 ± 0.3* 2.1 ± 0.3 <0.001 LVEDV (mL/m2) Rest 122 ± 18 99 ± 23* 133 ± 11† 0.007 Peak ex 124 ± 23 104 ± 26 134 ± 15 0.058 RVEDV (mL/m2) Rest 124 ± 22 81 ± 18* 125 ± 19† <0.001 Peak ex 119 ± 23 77 ± 20* 122 ± 22† <0.001 LVESV (mL/m2) Rest 59 ± 9† 52 ± 15 63 ± 17 0.272 Peak ex 44 ± 9 51 ± 22 58 ± 17 0.298 RVESV (mL/m2) Rest 59 ± 10 34 ± 10* 58 ± 2† <0.001 Peak ex 40 ± 9 26 ± 10 51 ± 17† 0.003 LVSV (mL/m2) Rest 63 ± 11 47 ± 10* 70 ± 9† 0.001 Peak ex 81 ± 15 53 ± 9* 76 ± 8† <0.001 RVSV (mL/m2) Rest 64 ± 11 47 ± 9* 67 ± 8† 0.001 Peak ex 79 ± 15 51 ± 10* 71 ± 7† <0.001 LVEF (%) Rest 51.1 ± 2.8 47.7 ± 5.5 53.0 ± 9.2 0.215 Peak ex 64.8 ± 3.7 52.8 ± 9.5* 57.2 ± 8.9 0.007 RVEF (%) Rest 51.9 ± 2.1 58.4 ± 6.6* 53.9 ± 3.3 0.017 Peak ex 66.3 ± 4.0 66.6 ± 5.9 59.0 ± 7.5 0.047 CI (L/min/m2) Rest 4.3 ± 1.4 3.1 ± 0.6 3.5 ± 0.5 0.070 Peak ex 12.3 ± 3.4 7.4 ± 1.6* 10.1 ± 1.1 0.001 LAVmax (mL/m2) Rest 52 ± 8 46 ± 15 57 ± 11 0.219 Peak ex 56 ± 18 53 ± 15 61 ± 20 0.745 RAVmax (mL/m2) Rest 78 ± 9 63 ± 26 95 ± 18† 0.020 Peak ex 68 ± 22 66 ± 29 76 ± 22 0.755 LAVmin (mL/m2) Rest 27 ± 5 24 ± 8 36 ± 10† 0.020 Peak ex 25 ± 8 31 ± 13 36 ± 11 0.202 RAVmin (mL/m2) Rest 42 ± 8 34 ± 15 59 ± 10*,† 0.003 Peak ex 26 ± 9 34 ± 18 40 ± 13 0.188 EA-healthy (n = 10) DCM (n = 9) EA-fibrosis (n = 5) P-value Heart rate (b.p.m.) Rest 66 ± 11 67 ± 8 52 ± 9*,† 0.025 Peak ex 153 ± 20 142 ± 13 138 ± 20 0.238 Mean arterial pressure (mmHg) Rest 88 ± 7 90 ± 10 81 ± 11 0.317 Peak ex 113 ± 14 117 ± 15 109 ± 7 0.580 Ea (mmHg/mL) Rest 1.9 ± 0.4 2.6 ± 0.6* 1.6 ± 0.4 0.003 Peak ex 2.1 ± 0.5 2.9 ± 0.3* 2.1 ± 0.3 <0.001 LVEDV (mL/m2) Rest 122 ± 18 99 ± 23* 133 ± 11† 0.007 Peak ex 124 ± 23 104 ± 26 134 ± 15 0.058 RVEDV (mL/m2) Rest 124 ± 22 81 ± 18* 125 ± 19† <0.001 Peak ex 119 ± 23 77 ± 20* 122 ± 22† <0.001 LVESV (mL/m2) Rest 59 ± 9† 52 ± 15 63 ± 17 0.272 Peak ex 44 ± 9 51 ± 22 58 ± 17 0.298 RVESV (mL/m2) Rest 59 ± 10 34 ± 10* 58 ± 2† <0.001 Peak ex 40 ± 9 26 ± 10 51 ± 17† 0.003 LVSV (mL/m2) Rest 63 ± 11 47 ± 10* 70 ± 9† 0.001 Peak ex 81 ± 15 53 ± 9* 76 ± 8† <0.001 RVSV (mL/m2) Rest 64 ± 11 47 ± 9* 67 ± 8† 0.001 Peak ex 79 ± 15 51 ± 10* 71 ± 7† <0.001 LVEF (%) Rest 51.1 ± 2.8 47.7 ± 5.5 53.0 ± 9.2 0.215 Peak ex 64.8 ± 3.7 52.8 ± 9.5* 57.2 ± 8.9 0.007 RVEF (%) Rest 51.9 ± 2.1 58.4 ± 6.6* 53.9 ± 3.3 0.017 Peak ex 66.3 ± 4.0 66.6 ± 5.9 59.0 ± 7.5 0.047 CI (L/min/m2) Rest 4.3 ± 1.4 3.1 ± 0.6 3.5 ± 0.5 0.070 Peak ex 12.3 ± 3.4 7.4 ± 1.6* 10.1 ± 1.1 0.001 LAVmax (mL/m2) Rest 52 ± 8 46 ± 15 57 ± 11 0.219 Peak ex 56 ± 18 53 ± 15 61 ± 20 0.745 RAVmax (mL/m2) Rest 78 ± 9 63 ± 26 95 ± 18† 0.020 Peak ex 68 ± 22 66 ± 29 76 ± 22 0.755 LAVmin (mL/m2) Rest 27 ± 5 24 ± 8 36 ± 10† 0.020 Peak ex 25 ± 8 31 ± 13 36 ± 11 0.202 RAVmin (mL/m2) Rest 42 ± 8 34 ± 15 59 ± 10*,† 0.003 Peak ex 26 ± 9 34 ± 18 40 ± 13 0.188 LAVmax, left atrial maximal volume; LAVmin, left atrial minimal volume; RAVmax, right atrial maximal volume; RAVmin, right atrial minimal volume; RVEDV, right ventricular end-diastolic volume. * P < 0.05 for difference vs. EA-healthy. †P < 0.05 for difference vs. DCM. At rest, LVEF was similar in the different groups. In the DCM and EA-fibrosis groups, less augmentation of LVEF (5 ± 6% vs. 4 ± 3% vs. 14 ± 3%; P = 0.001; Figure 2) and contractility [LVESPVR ratio; 1.4 (1.3–1.6) vs. 1.5 (1.3–1.6) vs. 1.8 (1.7–2.7); P < 0.001] was observed during exercise than EA-healthy. This was due to an attenuated reduction in LVESVi (Figure 3), whereas the response of LVEDVi was similar in all groups. Exercise-induced changes in Ea were similar between groups (+14 ± 23% vs. +13 ± 21% vs. 27 ± 8%; P = 0.422), whereas the change in LVESPVR was greater in EA-healthy compared with DCM and EA-fibrosis (+105 ± 51% vs. +38 ± 17% vs. +53 ± 18%). Figure 2 View largeDownload slide LVEF during exercise. All EA-healthy subjects demonstrate an increase in ΔLVEF during exercise as opposed to DCM patients and EA-fibrosis subjects in whom a heterogeneous response is observed (A). Asterisk indicates P-value <0.05 for the difference in ΔLVEF vs. EA-healthy. Error bars denote standard deviation. (B) End-systolic images at rest and at peak-exercise in a healthy EA, a patient with mild DCM and an EA with fibrosis. In the EA-healthy, LV and RV function augment with exercise. In contrast, LV function fails to augment in the DCM patient and the EA with fibrosis (right image). Figure 2 View largeDownload slide LVEF during exercise. All EA-healthy subjects demonstrate an increase in ΔLVEF during exercise as opposed to DCM patients and EA-fibrosis subjects in whom a heterogeneous response is observed (A). Asterisk indicates P-value <0.05 for the difference in ΔLVEF vs. EA-healthy. Error bars denote standard deviation. (B) End-systolic images at rest and at peak-exercise in a healthy EA, a patient with mild DCM and an EA with fibrosis. In the EA-healthy, LV and RV function augment with exercise. In contrast, LV function fails to augment in the DCM patient and the EA with fibrosis (right image). Figure 3 View largeDownload slide Changes in biventricular end-systolic volume during exercise. Changes in LVESVi (A) and RV ESVi (B) during incremental exercise are shown for EA-healthy (green), EA-fibrosis (black) and DCM patients (red). P-values are shown for the interaction between group and exercise-intensity. Error bars denote standard error of the mean. Figure 3 View largeDownload slide Changes in biventricular end-systolic volume during exercise. Changes in LVESVi (A) and RV ESVi (B) during incremental exercise are shown for EA-healthy (green), EA-fibrosis (black) and DCM patients (red). P-values are shown for the interaction between group and exercise-intensity. Error bars denote standard error of the mean. In a sub-analysis comparing EA-fibrosis and EA-healthy (n = 5), matched for exercise capacity (VO2peak 68.4 ± 4.2 vs. 64.0 ± 5.9 mL/kg/min; P = 0.210), the difference in LVEF reserve remained significant (ΔLVEF 4 ± 3% vs. 14 ± 2%; P < 0.001). DCM patients had higher resting RVEF compared with EA-healthy and EA-fibrosis subjects, the latter groups having similar values. As compared with EA-healthy, the exercise-induced increase in RVEF was diminished in both DCM and EA-fibrosis (ΔRVEF 8 ± 6% vs. 5 ± 5% vs. 14 ± 4%; P = 0.006) due to an attenuated reduction in RVESVi (P = 0.001 for interaction). Similar to LV functional measures, neither DCM nor EA-fibrosis demonstrated an increase in LAEF or RAEF during exercise, whereas LAEF and RAEF increased in EA-healthy (Figure 4). Figure 4 View largeDownload slide Changes in ventricular and atrial function during exercise. Changes in LVEF (A), RVEF (B) and LAEF (C) and RAEF (D) during incremental exercise are shown for EA-healthy (green), EA-fibrosis (black) and DCM patients (red). P-values are shown for the interaction between group and exercise-intensity. Error bars denote standard error of the mean. Figure 4 View largeDownload slide Changes in ventricular and atrial function during exercise. Changes in LVEF (A), RVEF (B) and LAEF (C) and RAEF (D) during incremental exercise are shown for EA-healthy (green), EA-fibrosis (black) and DCM patients (red). P-values are shown for the interaction between group and exercise-intensity. Error bars denote standard error of the mean. An example of exercise CMR comparing LV function in a DCM patient with EA-healthy and EA-fibrosis subjects is provided in Figure 2 and Supplementary data online, Video S1. Diagnostic accuracy to differentiate EA from DCM patients Receiver-operator characteristic curves demonstrated that cut-off values of 11.2% for the increase in LVEF from rest to peak exercise (AUC = 0.92, P < 0.001) had a sensitivity of 93% and specificity of 90% to differentiate EA-healthy from DCM and EA-fibrosis, whereas the LVESPVR ratio of 1.8 (AUC = 0.94; P = 0.003) had a sensitivity of 83% and a specificity of 100% patients; Figure 5). In contrast, resting LVEF was not predictive (AUC = 0.56, P = 0.598). Similarly, VO2peak was unable to accurately separate the cohorts (AUC = 0.68, P = 0.128). The AUC value for ΔLVEF was statistically different from that of resting LVEF (P = 0.013) and similar to that of the LVESPVR ratio (P = 0.879). Figure 5 View largeDownload slide Differentiation between physiological and pathological left ventricular remodelling. Receiver operating curves for the ability of resting and exercise measurements of LVEF and ΔLVEF, peak oxygen consumption (VO2peak), and contractility (end-systolic pressure–volume ratio) to differentiate healthy EAs with physiological remodelling from DCM patients and EAs with fibrosis. Figure 5 View largeDownload slide Differentiation between physiological and pathological left ventricular remodelling. Receiver operating curves for the ability of resting and exercise measurements of LVEF and ΔLVEF, peak oxygen consumption (VO2peak), and contractility (end-systolic pressure–volume ratio) to differentiate healthy EAs with physiological remodelling from DCM patients and EAs with fibrosis. Discussion The main goal of this study was to determine whether exercise evaluation of cardiac reserve enables differentiation between physiological adaptation to endurance exercise and pathological LV remodelling. Whereas LV systolic function was similar between athletes and DCM at rest, exercise imaging reliably distinguished between healthy athletes and those with pathology (DCM and athletes with LV fibrosis). Importantly, by comparing healthy athletes to elite EA with manifest LV fibrosis, we demonstrated that exercise capacity alone does not exclude significant LV damage. Therefore, evaluation of contractile reserve may be a useful tool, which can be applied in both athletic and non-athletic populations to separate health from disease. Although multiple studies reported profound increases in LV and RV mass and volumes due to intensive endurance training, significant variability exists in the extent of cardiac remodelling. In a study on Tour de France cyclists, Abergel et al.1 reported that about one-half of athletes had substantial LV enlargement and 12% had an LVEF <52%, highlighting the diagnostic overlap between health and DCM. Accurate differentiation between these entities is of paramount importance. Incorrectly labelling an athlete with pathology could lead to inappropriate exclusion from sport and, of even greater consequence, failing to identify early pathology will mean that treatments with proven prognostic benefit (e.g. ACE inhibitors) may be delayed, and the patient may be exposed to an increased risk of cardiomyopathy and arrhythmic sudden cardiac death.2,18 Current diagnostic tests rely on resting assessments and are unproven. Similarly, temporal cessation of sports activity (‘detraining’) incurs significant periods of uncertainty for the athlete and has not been shown to assist in decision-making. Given that athletic ventricular enlargement has been observed to persist for many years,19 it is unlikely that short-term detraining could reliably distinguish athletic remodelling from pathology. Our current results demonstrate that evaluation during exercise facilitates the differentiation between athlete’s heart and pathological LV adaptation with considerable accuracy. In both groups with LV pathology, there was impaired contractile reserve characterized by failure to decrease end-systolic volumes and increase LVEF.20 In contrast, healthy EAs demonstrated a consistent reduction in LV end-systolic volumes and a significant increase in LVEF, irrespective of resting LVEF. Although criticism is often levelled at the load-dependent volume measures and EF, we found no advantage in combining these measures with invasively derived pressures. The most direct assessment of contractility would be to adjust volumes for pressure, and thus, we assessed the LV end-systolic pressure–volume relationship as a surrogate of contractility. DCM and EA-fibrosis subjects had similar changes in arterial elastance but reduced contractile reserve. Taken together, these findings suggest exercise-induced uncoupling between the LV and the arterial system, as evidenced by impaired augmentation of LVEF, a non-invasive surrogate of ventricular-arterial coupling.21 Importantly, we demonstrated that purely non-invasive volumetric measures proved as accurate in separating health from disease. This provides obvious advantages for the future utility of these techniques. Future work will have to determine whether exercise echocardiography is as reliable as our CMR evaluation, although image quality and through-plane movement during exercise provide limitations in feasibility and reproducibility.8 An important finding of this study is that maximal peak oxygen consumption was paradoxically higher in athletes with LV damage compared with healthy EA, while peak CI was similar in both groups. In combination, these findings suggest that peripheral factors play an important role in the preservation of cardiopulmonary exercise performance despite a reduction in LV contractile reserve. This finding is in keeping with previous data in hypertrophic cardiomyopathy in which exercise testing facilitated the diagnosis in only one-fifth of subjects.6 The EA with fibrosis in the current study were high-level competitive athletes as evidenced by their excellent exercise capacity. Hence, all of them had a VO2peak >60 mL/kg/min. Nevertheless, they all presented with ventricular arrhythmias, either at initial workup or during follow-up, as previously reported, indicating that superb exercise capacity per se does not imply a good prognosis.4 Another important finding of this study is that the functional abnormalities during exercise were not restricted to the LV, but also involved the RV and both atria. Previous studies using radionuclide ventriculography or invasive techniques have assessed either the LV16 or the RV22 response to exercise in severe DCM. In addition, several studies used pharmacological stress echocardiography or stress CMR to assess LV or biventricular contractile reserve.23–25 To date, however, no studies have evaluated the response of both ventricles and atria to incremental exercise simultaneously. In keeping with previous data, we found that the increase in RVEF was diminished in the DCM and particularly the EA-fibrosis group compared with healthy EAs. Furthermore, changes in bi-atrial function during exercise mirrored the changes observed in the ventricles, as evidenced by attenuated increases in LA and RA emptying function. This is in line with previous observations revealing that the degree of left atrial dysfunction at rest is greater than expected from the degree of atrial dilatation, thereby suggesting myopathy of the left atrium.26 Moreover, there is data to suggest that left atrial myopathy may even precede LV myopathy and can be used for early detection and risk stratification in some cases.27,28 In this context, the normal functional response of both atria to exercise in EAs in the current study provides additional reassurance that the low resting LVEF in EAs is not due to underlying myocardial pathology, but rather represents physiological remodelling. The findings of this study have clinical implications for subjects at risk for the development of DCM, e.g. asymptomatic family members of patients with severe DCM or prior myocarditis. Due to the design of this study, comparing contractile reserve in EAs and patients with only mild DCM, the degree of resting LV impairment in the DCM group was substantially less profound than in previous studies evaluating contractile reserve in DCM.14,22,23 As a result, exercise capacity was preserved, NTproBNP was only mildly elevated and the diagnosis of mild DCM was mainly based on a mildly reduced resting LVEF and a high index of suspicion due to disease in other family members. As such, these subjects were well suited for comparison with the EAs because the differentiation from Athlete’s Heart would be clinically obvious in patients with severe DCM. Until present, there had been no studies evaluating LV contractile reserve in this subset of DCM patients with mild phenotype despite the recommendation to perform exercise imaging as part of the diagnostic workup.7 We observed that LV functional reserve in the DCM patients was heterogeneous and independent of resting LVEF. This is consistent with previous data in patients with severe DCM,14 which suggests that LV contractile reserve may improve clinical risk stratification and guide therapeutic management of these subjects in the early course of the disease. Further studies are required to verify whether those subjects with higher contractile reserve have better outcome as compared with those in whom cardiac function deteriorates during exercise. Limitations Firstly, the comprehensive measurements undertaken in this proof-of-concept study and the unique characteristics of the cohorts limited the sample size. The small sample size may have increased the probability of Type II statistical errors while multiple comparisons increase the likelihood of Type I errors. Nevertheless, to date this is the only study to compare exercise measures in EAs and DCM patients. It would have been valuable to include a control group of athletes with established DCM. However, given the low prevalence of DCM amongst athletes, we included a cohort of elite EA with underlying LV damage associated with ventricular arrhythmias. This is the largest cohort to date to assess cardiac function and exercise metrics. The established accuracy of exercise CMR measures enabled us to evaluate meaningful haemodynamic differences even within this modest-sized cohort. The fact that highly significant differences were apparent reinforces the accuracy of ex-CMR and the extent of the physiological differences. Further larger and prospective studies are required to validate the cut-off values reported in this study. LVESPVR is not equal to end-systolic elastance. Calculation of V0 typically requires the use of invasively derived pressure–volume loops, which was considered far too invasive for the scope of this study. Nevertheless, LVESPVR is considered a valid approximation of end-systolic elastance.11,29 We cannot exclude that V0 changes during exercise. However, previous studies reported that V0 remains unaltered during exercise or changes in loading conditions.30,31 As expected, the majority of DCM patients had received treatment including beta-blockers or calcium blockers. This introduces confounders in comparisons with the EAs and control subjects. However, beta-blocker and calcium channel blocking medications were withheld for at least 24 h prior to exercise testing; a period sufficient to exclude any persisting pharmacodynamic effect. Finally, although the cardiac response to exercise in the EAs was similar to previous observations in healthy athletes,9 longer follow-up is necessary to assess whether the mildly reduced resting LV function is associated with clinical events, e.g. arrhythmias, in the long term. Similarly, larger studies are required to verify whether assessment of contractile reserve provides additional value for risk stratification in DCM patients with mildly reduced LVEF. Conclusions Neither exercise capacity nor resting measures of cardiac function are helpful in differentiating healthy athletes with borderline LVEF from those with myocardial impairment or fibrosis. On the other hand, augmentation of biventricular and atrial function during exercise are important tools that can accurately differentiate between physiological and pathological LV remodelling. Supplementary data Supplementary data are available at European Heart Journal - Cardiovascular Imaging online. Funding This study was funded by a grant from the Fund for Scientific Research Flanders (FWO), Belgium. Andre La Gerche was funded by a Career Development Fellowship from the National Health and Medical Research Council and National Heart Foundation of Australia (NHMRC and NHF). Frédéric Schnell was funded by a grant of the French Federation of Cardiology. Conflict of interest: None declared. References 1 Abergel E , Chatellier G , Hagege AA , Oblak A , Linhart A , Ducardonnet A et al. Serial left ventricular adaptations in world-class professional cyclists: implications for disease screening and follow-up . J Am Coll Cardiol 2004 ; 44 : 144 – 9 . Google Scholar CrossRef Search ADS PubMed 2 Maron BJ , Doerer JJ , Haas TS , Tierney DM , Mueller FO. Sudden deaths in young competitive athletes: analysis of 1866 deaths in the United States, 1980-2006 . Circulation 2009 ; 119 : 1085 – 92 . Google Scholar CrossRef Search ADS PubMed 3 Heidbuchel H , Hoogsteen J , Fagard R , Vanhees L , Ector H , Willems R et al. High prevalence of right ventricular involvement in endurance athletes with ventricular arrhythmias. Role of an electrophysiologic study in risk stratification . Eur Heart J 2003 ; 24 : 1473 – 80 . Google Scholar CrossRef Search ADS PubMed 4 Schnell F , Claessen G , La Gerche A , Bogaert J , Lentz PA , Claus P et al. Subepicardial delayed gadolinium enhancement in asymptomatic athletes: let sleeping dogs lie? Br J Sports Med 2016 ; 50 : 111 – 7 . Google Scholar CrossRef Search ADS PubMed 5 Trivax JE , McCullough PA. Phidippides cardiomyopathy: a review and case illustration . Clin Cardiol 2012 ; 35 : 69 – 73 . Google Scholar CrossRef Search ADS PubMed 6 Sheikh N , Papadakis M , Schnell F , Panoulas V , Malhotra A , Wilson M et al. Clinical profile of athletes with hypertrophic cardiomyopathy . Circ Cardiovasc Imaging 2015 ; 8 : e003454. Google Scholar CrossRef Search ADS PubMed 7 Galderisi M , Cardim N , D'Andrea A , Bruder O , Cosyns B , Davin L et al. The multi-modality cardiac imaging approach to the Athlete's heart: an expert consensus of the European Association of Cardiovascular Imaging . Eur Heart J Cardiovasc Imaging 2015 ; 16 : 353. Google Scholar CrossRef Search ADS PubMed 8 Claessen G , La Gerche A , Voigt JU , Dymarkowski S , Schnell F , Petit T et al. Accuracy of echocardiography to evaluate pulmonary vascular and RV function during exercise . JACC Cardiovasc Imaging 2016 ; 9 : 532 – 43 . Google Scholar CrossRef Search ADS PubMed 9 La Gerche A , Claessen G , Dymarkowski S , Voigt JU , De Buck F , Vanhees L et al. Exercise-induced right ventricular dysfunction is associated with ventricular arrhythmias in endurance athletes . Eur Heart J 2015 ; 36 : 1998 – 2010 . Google Scholar CrossRef Search ADS PubMed 10 La Gerche A , Claessen G , Van de Bruaene A , Pattyn N , Van Cleemput J , Gewillig M et al. Cardiac MRI: a new gold standard for ventricular volume quantification during high-intensity exercise . Circ Cardiovasc Imaging 2013 ; 6 : 329 – 38 . Google Scholar CrossRef Search ADS PubMed 11 Kelly RP , Ting CT , Yang TM , Liu CP , Maughan WL , Chang MS et al. Effective arterial elastance as index of arterial vascular load in humans . Circulation 1992 ; 86 : 513 – 21 . Google Scholar CrossRef Search ADS PubMed 12 Pellicori P , Zhang J , Lukaschuk E , Joseph AC , Bourantas CV , Loh H et al. Left atrial function measured by cardiac magnetic resonance imaging in patients with heart failure: clinical associations and prognostic value . Eur Heart J 2015 ; 36 : 733 – 42 . Google Scholar CrossRef Search ADS PubMed 13 Hoit BD. Left atrial size and function: role in prognosis . J Am Coll Cardiol 2014 ; 63 : 493 – 505 . Google Scholar CrossRef Search ADS PubMed 14 Kirlin PC , Das S , Zijnen P , Wijns W , Domenicucci S , Roelandt J et al. The exercise response in idiopathic dilated cardiomyopathy . Clin Cardiol 1984 ; 7 : 205 – 10 . Google Scholar CrossRef Search ADS PubMed 15 Konishi T , Koyama T , Aoki T , Yada T , Futagami Y , Sekioka K et al. Assessments of left ventricular function during exercise in patients with dilated cardiomyopathy: comparison with ischemic cardiomyopathy . J Cardiol 1989 ; 19 : 797 – 804 . Google Scholar PubMed 16 Metz CE. Basic principles of ROC analysis . Semin Nucl Med 1978 ; 8 : 283 – 98 . Google Scholar CrossRef Search ADS PubMed 17 DeLong ER , DeLong DM , Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach . Biometrics 1988 ; 44 : 837 – 45 . Google Scholar CrossRef Search ADS PubMed 18 Corrado D , Basso C , Rizzoli G , Schiavon M , Thiene G. Does sports activity enhance the risk of sudden death in adolescents and young adults? J Am Coll Cardiol 2003 ; 42 : 1959 – 63 . Google Scholar CrossRef Search ADS PubMed 19 Pelliccia A , Maron BJ , De Luca R , Di Paolo FM , Spataro A , Culasso F. Remodeling of left ventricular hypertrophy in elite athletes after long-term deconditioning . Circulation 2002 ; 105 : 944 – 9 . Google Scholar CrossRef Search ADS PubMed 20 Haddad F , Vrtovec B , Ashley EA , Deschamps A , Haddad H , Denault AY. The concept of ventricular reserve in heart failure and pulmonary hypertension: an old metric that brings us one step closer in our quest for prediction . Curr Opin Cardiol 2011 ; 26 : 123 – 31 . Google Scholar CrossRef Search ADS PubMed 21 Cohen-Solal A , Faraggi M , Czitrom D , Le Guludec D , Delahaye N , Gourgon R. Left ventricular-arterial system coupling at peak exercise in dilated nonischemic cardiomyopathy . Chest 1998 ; 113 : 870 – 7 . Google Scholar CrossRef Search ADS PubMed 22 Plehn G , Vormbrock J , Perings S , Plehn A , Meissner A , Butz T et al. Comparison of right ventricular functional response to exercise in hypertrophic versus idiopathic dilated cardiomyopathy . Am J Cardiol 2010 ; 105 : 116 – 21 . Google Scholar CrossRef Search ADS PubMed 23 Matsumoto K , Tanaka H , Onishi A , Motoji Y , Tatsumi K , Sawa T et al. Bi-ventricular contractile reserve offers an incremental prognostic value for patients with dilated cardiomyopathy . Eur Heart J Cardiovasc Imaging 2015 ; 16 : 1213 – 23 . Google Scholar CrossRef Search ADS PubMed 24 Pratali L , Picano E , Otasevic P , Vigna C , Palinkas A , Cortigiani L et al. Prognostic significance of the dobutamine echocardiography test in idiopathic dilated cardiomyopathy . Am J Cardiol 2001 ; 88 : 1374 – 8 . Google Scholar CrossRef Search ADS PubMed 25 Naqvi TZ , Goel RK , Forrester JS , Siegel RJ. Myocardial contractile reserve on dobutamine echocardiography predicts late spontaneous improvement in cardiac function in patients with recent onset idiopathic dilated cardiomyopathy . J Am Coll Cardiol 1999 ; 34 : 1537 – 44 . Google Scholar CrossRef Search ADS PubMed 26 Triposkiadis F , Pitsavos C , Boudoulas H , Trikas A , Toutouzas P. Left atrial myopathy in idiopathic dilated cardiomyopathy . Am Heart J 1994 ; 128 : 308 – 15 . Google Scholar CrossRef Search ADS PubMed 27 Graber HL , Unverferth DV , Baker PB , Ryan JM , Baba N , Wooley CF. Evolution of a hereditary cardiac conduction and muscle disorder: a study involving a family with six generations affected . Circulation 1986 ; 74 : 21 – 35 . Google Scholar CrossRef Search ADS PubMed 28 Otway R , Vandenberg JI , Guo G , Varghese A , Castro ML , Liu J et al. Stretch-sensitive KCNQ1 mutation A link between genetic and environmental factors in the pathogenesis of atrial fibrillation? J Am Coll Cardiol 2007 ; 49 : 578 – 86 . Google Scholar CrossRef Search ADS PubMed 29 Sagawa K , Suga H , Shoukas AA , Bakalar KM. End-systolic pressure/volume ratio: a new index of ventricular contractility . Am J Cardiol 1977 ; 40 : 748 – 53 . Google Scholar CrossRef Search ADS PubMed 30 Little WC , Cheng CP , Peterson T , Vinten-Johansen J. Response of the left ventricular end-systolic pressure-volume relation in conscious dogs to a wide range of contractile states . Circulation 1988 ; 78 : 736 – 45 . Google Scholar CrossRef Search ADS PubMed 31 Maughan WL , Sunagawa K , Burkhoff D , Sagawa K. Effect of arterial impedance changes on the end-systolic pressure-volume relation . Circ Res 1984 ; 54 : 595 – 602 . Google Scholar CrossRef Search ADS PubMed Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2018. For permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png European Heart Journal – Cardiovascular Imaging Oxford University Press

Loading next page...
 
/lp/ou_press/exercise-cardiac-magnetic-resonance-to-differentiate-athlete-s-heart-pC6e61KILI
Publisher
Oxford University Press
Copyright
Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2018. For permissions, please email: journals.permissions@oup.com.
ISSN
2047-2404
D.O.I.
10.1093/ehjci/jey050
Publisher site
See Article on Publisher Site

Abstract

Abstract Aims The distinction between left ventricular (LV) dilation with mildly reduced LV ejection fraction (EF) in response to regular endurance exercise training and an early cardiomyopathy is a frequently encountered and difficult clinical conundrum. We hypothesized that exercise rather than resting measures would provide better discrimination between physiological and pathological LV remodelling and that preserved exercise capacity does not exclude significant LV damage. Methods and results We prospectively included 19 subjects with LVEF between 40 and 52%, comprising 10 ostensibly healthy endurance athletes (EA-healthy) and nine patients with dilated cardiomyopathy (DCM). In addition, we recruited five EAs with a region of subepicardial LV. Receiver operating characteristic fibrosis (EA-fibrosis). Cardiac magnetic resonance (CMR) imaging was performed at rest and during supine bicycle exercise. Invasive afterload measures were obtained to enable calculations of biventricular function relative to load (an estimate of contractility). In DCM and EA-fibrosis subjects there was diminished augmentation of LVEF (5 ± 6% vs. 4 ± 3% vs. 14 ± 3%; P = 0.001) and contractility [LV end-systolic pressure–volume ratio, LVESPVR; 1.4 (1.3–1.6) vs. 1.5 (1.3–1.6) vs. 1.8 (1.7–2.7); P < 0.001] during exercise relative to EA-healthy. Receiver-operator characteristic curves demonstrated that a cut-off value of 11.2% for ΔLVEF differentiated DCM and EA-fibrosis patients from EA-healthy [area under the curve (AUC) = 0.92, P < 0.001], whereas resting LVEF and VO2max were not predictive. The AUC value for LVESPVR ratio was similar to that of ΔLVEF. Conclusions Functional cardiac evaluation during exercise is a promising tool in differentiating healthy athletes with borderline LVEF from those with an underlying cardiomyopathy. Excellent exercise capacity does not exclude significant LV damage. athlete’s heart , dilated cardiomyopathy , exercise , myocardial fibrosis , cardiac magnetic resonance imaging , contractile reserve Introduction Profound left ventricular (LV) dilation with mildly reduced LV ejection fraction (EF) at rest is seen in >10% of elite endurance athletes (EA).1 Distinguishing this phenotype from true underlying LV systolic dysfunction can be a difficult clinical conundrum1 with important consequences for competitive sport participation.2 The two conditions are sometimes considered mutually exclusive based on the assumption that excellent exercise capacity precludes significant pathology. However, we and others have described athletes with significant pathology diagnosed whilst continuing to compete at the highest level.3–6 Thus, new methods are required to accurately separate health from pathology amongst athletes with ambiguous cardiac function at rest. Guidelines suggest that exercise assessments of cardiac function can be used to differentiate athlete’s heart from cardiomyopathy when resting measures of function appear abnormal.7 Although a logical premise, there have not been any studies to support this assertion and the accuracy of exercise imaging is uncertain. We have previously demonstrated that real-time cardiac magnetic resonance imaging performed during exercise (ex-CMR) can be used to accurately quantify cardiac function and represents the ideal modality for investigating differences in contractile reserve between normal and diseased myocardium.8–10 This study sought to evaluate the cardiac response to dynamic exercise in ostensibly healthy EA with mildly reduced LVEF as compared with a group of non-athletic subjects with mild or early dilated cardiomyopathy (DCM). Furthermore, we tested whether ex-CMR was useful to identify sub-clinical cardiac damage. Therefore, we included a third group of highly trained athletes in whom CMR had incidentally detected significant LV fibrosis while they were still competing at an elite level with normal exercise capacity. We hypothesized that LV contractile reserve would be able to discriminate between healthy athletes and apparently normal athletic subjects with underlying LV damage. Methods Subjects Thirteen EA (all male) with low normal resting LVEF were recruited using a prospective inclusion criterion of LVEF ≤52% measured by CMR, in accordance with previous studies.1 These EA-healthy athletes were recruited from volunteers responding to advertisements at local triathlon and cycling clubs (n = 4) and from individuals referred to our institution (n = 9) after screening examinations revealed a mildly reduced LV systolic function in the absence of other signs of structural heart disease (i.e. normal wall thickness, no valvular heart disease, no regional wall motion abnormalities, and no signs of myocardial fibrosis). EA-healthy subjects were included if: (i) they were participating in regular cycling and/or running training of >6 h/week, (ii) they had no history of cardiovascular disease, and (iii) CMR at inclusion confirmed an LVEF ≤52%. Three athletes were excluded because CMR at enrolment revealed a normal LVEF prior to study participation. Therefore, the final EA-healthy cohort consisted of 10 subjects. Nine patients with mild DCM (eight male) and five EA with fibrosis (EA-fibrosis, all male) were recruited from an existing database in addition to new cases presenting over the study period. The DCM cohort consisted of seven first-degree family members of DCM patients without an identifiable mutation and two patients with partially recovered severe DCM. All subjects had LVEF ≥40 and ≤52% measured by CMR prior to study participation. The cohort of EA-fibrosis subjects consisted of high-level EA in whom significant delayed gadolinium enhancement (11.4 ± 4.7% of LV mass) was detected after screening evaluation revealed pathological T-wave inversion or ventricular arrhythmias, as previously described.4 Relative to our previous study cohort,4 only those EA-fibrosis subjects referred to our institution were included in this study, as well as two new cases presenting during the study period. Therefore, the final cohort consisted of five EA-fibrosis subjects. An example of an EA-fibrosis subject is shown in Figure 1. All of the EA-fibrosis subjects developed ventricular arrhythmias resulting in exclusion from competitive sports participation. Figure 1 View largeDownload slide Example of an endurance athlete with significant subepicardial delayed gadolinium enhancement (arrowheads). Figure 1 View largeDownload slide Example of an endurance athlete with significant subepicardial delayed gadolinium enhancement (arrowheads). To ascertain whether significant LV damage and elite exercise capacity could coexist, contractile reserve in the EA-fibrosis cohort was compared against a subset of EA-healthy subjects, matched for exercise capacity. The study protocol conformed to the Declaration of Helsinki and was approved by the Ethics Committee of UZ Leuven (B322201214035). All subjects provided informed consent. Study design Prior to CMR evaluation, a 20-gauge arterial catheter was placed in the radial artery for the measurement of systemic arterial pressures. Subsequently, biventricular volumes were measured during supine cycling exercise using a real-time CMR method that we previously described and validated against invasive standards.10 In brief, subjects performed supine exercise within the CMR bore using a cycle ergometer with adjustable electronic resistance (Lode, Groningen, The Netherlands). Images were acquired using a Philips Achieva 1.5 T CMR with a five-element phased-array coil (Philips Medical Systems, Best, The Netherlands) at rest and during supine bicycle exercise at 25%, 50%, and 66% of maximal power determined by previous upright cardiopulmonary exercise testing.10 Steady-state free precession cine imaging was performed without cardiac gating. Imaging parameters were: field of view 320 × 260 mm (approximately), 128 × 128 matrix, flip angle 50°, SENSE factor 2 (Cartesian k-space under-sampling), repetition time 1.8 ms, echo time 0.9 ms, and reconstructed voxel size 2.3 × 2.3 × 8 mm. A 3D stack of 13–18 contiguous 8 mm image slices, covering both ventricles from apex to base, was serially acquired in the short-axis plane and subsequently in the horizontal long-axis plane. All image frames were acquired during free breathing with a temporal resolution of 36–38 ms. Systemic arterial pressure measurements were continuously recorded during the exercise CMR protocol and analysed off-line using LabChart v6.1.1 (AD Instruments). Using in-house developed software (RightVol, Leuven, Belgium), LV and right ventricular (RV) end-diastolic volumes and end-systolic volumes (EDVi, ESVi) and left atrial and right atrial (LA, RA) maximal volumes and minimal volumes (Vmax, Vmin) were calculated by a summation of disks and indexed for body surface area. LVEF and RVEF were calculated as (EDVi–ESVi)/EDVi. Stroke volume was measured as EDVi – ESVi and cardiac index (CI) as the product of SVi and heart rate. If assuming that V0 (zero-volume intercept of the end-systolic pressure–volume relationship) is negligible, a single point LV end-systolic pressure–volume ratio (LVESPVR) can be calculated using the formula (0.9*systolic blood pressure)/LVESV as a surrogate of ventricular elastance.9,11 LV contractile reserve was defined as a ratio of peak-exercise to resting LVESPVR (subsequently referred to as ‘LVESPVR ratio’). Arterial elastance (Ea) was calculated as (0.9*systolic blood pressure)/LVSV.11 As a measure of global atrial function, atrial total emptying fraction (LAEF, RAEF) was calculated as [(Vmax – Vmin)/Vmax].12,13 NT-proBNP was analysed from venous blood samples. Statistical analysis Data were analysed using IBM SPSS statistics 22 software. Gaussian distribution of all continuous variables was tested using a Kolmogorov–Smirnov test. Descriptive data for continuous variables are presented as mean ± standard deviation or as medians (25% and 75% percentile) as appropriate. Comparisons between groups for continuous variables were performed by one-way analysis of variance (ANOVA) or the Kruskal–Wallis with Bonferroni or Dunn’s test for multiple comparison posthoc correction, as appropriate. The Fisher’s exact or the χ2 test was used for categorical variables. To determine the sample sizes, the following estimates were used: in a previous study using exercise CMR, we demonstrated that healthy EA had an 8 ± 6% increase in LVEF from rest to maximal exercise.9 According to our hypothesis, we predicted that LVEF will not change (0% increase) during exercise in the subjects with underlying LV damage.14,15 Using these assumptions, a sample size of n = 9 was calculated to provide 80% power in detecting impaired LVEF augmentation during exercise in the groups with LV pathology (α = 5%, 1−β = 80%, n = 9). The biventricular volume response from rest to peak-intensity exercise in the different groups was compared using repeated measures ANOVA with exercise-intensity as within-subject effect and group (DCM vs. EA-fibrosis vs. EA-healthy) as a between-subject effect. Receiver operating characteristic (ROC) curves were constructed to determine the diagnostic accuracy of resting and exercise measures of LV function16 for distinguishing EA-healthy subjects from both non-athletic and athletic subjects with underlying LV damage (expressed as area and 95% confidence intervals). The ‘optimal’ cut-off value for each parameter was defined as the value of the parameter that corresponded with the highest sum of specificity and sensitivity. The significance of differences in area under the curve (AUC) of the correlated rest and peak exercise ROC curves was tested using the methodology described by DeLong et al.17 A P-value <0.05 was considered statistically significant. Results The demographic, clinical characteristics, and cardiopulmonary exercise testing data are presented in Table 1. All groups were of similar age and gender. As expected, both EA-healthy and EA-fibrosis cohorts had superior exercise capacity compared with DCM patients. A majority of DCM patients received therapy with a beta-blocker and angiotensin converting enzyme (ACE) inhibitor. All negative chronotropic medication was withheld for at least 24 h prior to exercise testing. NT-proBNP tended to be higher in DCM patients compared with the other groups. Invasive pressure measurement was performed in 20 of 24 study participants (6/10 EA-healthy, 9/9 DCM, and 5/5 EA-fibrosis). All EA-healthy subjects and DCM patients had resting LVEF <55% at the time of the exercise CMR protocol (noting some variability from the measure used for study inclusion). All of the EA-fibrosis subjects had significant delayed gadolinium enhancement (11.4 ± 4.7% of LV mass) as compared with two DCM subjects and none of the EA-healthy subjects. One of the EA-fibrosis subjects also had delayed gadolinium enhancement (DGE) of the RV free wall. Table 1 Baseline characteristics EA-healthy (n = 10) DCM (n = 9) EA-fibrosis (n = 5) ANOVA P-value Clinical  Age (years) 34 ± 12 44 ± 14 33 ± 8 0.149  BMI (kg/m2) 23.8 ± 2.3 25.1 ± 3.8 22.0 ± 0.7 0.173  Weight (kg) 77.9 ± 9.2 78.2 ± 16.2 78.1 ± 6.0 0.998  Male, n 9 8 5 1.00 Medication  Beta-blockers, n 0 6 1 0.005  ACE inhibitors, n 0 6 0 0.001  AR blockers, n 0 2 0 0.162 Biochemical  NT-proBNP (pg/mL) 32 (14–49) 146 (51–320) 57 (19–83) 0.062 CPET  VO2peak (mL/min) 4343 ± 760 2253 ± 534* 5343 ± 532*,† <0.001  VO2peak (mL/min/kg) 56.2 ± 10.1 29.0 ± 5.3* 68.4 ± 4.2*,† <0.001  Peak power (watts) 389 ± 70 206 ± 55* 444 ± 56† <0.001  Peak HR (b.p.m.) 182 ± 8 164 ± 19* 176 ± 6 0.031 CMR at inclusion  LVEF (%) 49.7 ± 1.9 47.7 ± 5.5 53.0 ± 9.2 0.243  DGE, n 0 2 5 <0.001 EA-healthy (n = 10) DCM (n = 9) EA-fibrosis (n = 5) ANOVA P-value Clinical  Age (years) 34 ± 12 44 ± 14 33 ± 8 0.149  BMI (kg/m2) 23.8 ± 2.3 25.1 ± 3.8 22.0 ± 0.7 0.173  Weight (kg) 77.9 ± 9.2 78.2 ± 16.2 78.1 ± 6.0 0.998  Male, n 9 8 5 1.00 Medication  Beta-blockers, n 0 6 1 0.005  ACE inhibitors, n 0 6 0 0.001  AR blockers, n 0 2 0 0.162 Biochemical  NT-proBNP (pg/mL) 32 (14–49) 146 (51–320) 57 (19–83) 0.062 CPET  VO2peak (mL/min) 4343 ± 760 2253 ± 534* 5343 ± 532*,† <0.001  VO2peak (mL/min/kg) 56.2 ± 10.1 29.0 ± 5.3* 68.4 ± 4.2*,† <0.001  Peak power (watts) 389 ± 70 206 ± 55* 444 ± 56† <0.001  Peak HR (b.p.m.) 182 ± 8 164 ± 19* 176 ± 6 0.031 CMR at inclusion  LVEF (%) 49.7 ± 1.9 47.7 ± 5.5 53.0 ± 9.2 0.243  DGE, n 0 2 5 <0.001 BMI, body mass index, AR, angiotensin II receptor; CPET, cardiopulmonary exercise testing; DGE, delayed gadolinium enhancement. * P < 0.05 for difference vs. EA-healthy. †P < 0.05 for difference vs. DCM. Table 1 Baseline characteristics EA-healthy (n = 10) DCM (n = 9) EA-fibrosis (n = 5) ANOVA P-value Clinical  Age (years) 34 ± 12 44 ± 14 33 ± 8 0.149  BMI (kg/m2) 23.8 ± 2.3 25.1 ± 3.8 22.0 ± 0.7 0.173  Weight (kg) 77.9 ± 9.2 78.2 ± 16.2 78.1 ± 6.0 0.998  Male, n 9 8 5 1.00 Medication  Beta-blockers, n 0 6 1 0.005  ACE inhibitors, n 0 6 0 0.001  AR blockers, n 0 2 0 0.162 Biochemical  NT-proBNP (pg/mL) 32 (14–49) 146 (51–320) 57 (19–83) 0.062 CPET  VO2peak (mL/min) 4343 ± 760 2253 ± 534* 5343 ± 532*,† <0.001  VO2peak (mL/min/kg) 56.2 ± 10.1 29.0 ± 5.3* 68.4 ± 4.2*,† <0.001  Peak power (watts) 389 ± 70 206 ± 55* 444 ± 56† <0.001  Peak HR (b.p.m.) 182 ± 8 164 ± 19* 176 ± 6 0.031 CMR at inclusion  LVEF (%) 49.7 ± 1.9 47.7 ± 5.5 53.0 ± 9.2 0.243  DGE, n 0 2 5 <0.001 EA-healthy (n = 10) DCM (n = 9) EA-fibrosis (n = 5) ANOVA P-value Clinical  Age (years) 34 ± 12 44 ± 14 33 ± 8 0.149  BMI (kg/m2) 23.8 ± 2.3 25.1 ± 3.8 22.0 ± 0.7 0.173  Weight (kg) 77.9 ± 9.2 78.2 ± 16.2 78.1 ± 6.0 0.998  Male, n 9 8 5 1.00 Medication  Beta-blockers, n 0 6 1 0.005  ACE inhibitors, n 0 6 0 0.001  AR blockers, n 0 2 0 0.162 Biochemical  NT-proBNP (pg/mL) 32 (14–49) 146 (51–320) 57 (19–83) 0.062 CPET  VO2peak (mL/min) 4343 ± 760 2253 ± 534* 5343 ± 532*,† <0.001  VO2peak (mL/min/kg) 56.2 ± 10.1 29.0 ± 5.3* 68.4 ± 4.2*,† <0.001  Peak power (watts) 389 ± 70 206 ± 55* 444 ± 56† <0.001  Peak HR (b.p.m.) 182 ± 8 164 ± 19* 176 ± 6 0.031 CMR at inclusion  LVEF (%) 49.7 ± 1.9 47.7 ± 5.5 53.0 ± 9.2 0.243  DGE, n 0 2 5 <0.001 BMI, body mass index, AR, angiotensin II receptor; CPET, cardiopulmonary exercise testing; DGE, delayed gadolinium enhancement. * P < 0.05 for difference vs. EA-healthy. †P < 0.05 for difference vs. DCM. Cardiac response to exercise Resting and peak exercise cardiac haemodynamics are shown in Table 2. LV and RV end-diastolic volumes and RV end-systolic volumes were smaller in DCM compared with the EA-healthy and EA-fibrosis, whereas LV end-systolic volumes were similar. Table 2 Cardiac haemodynamics EA-healthy (n = 10) DCM (n = 9) EA-fibrosis (n = 5) P-value Heart rate (b.p.m.) Rest 66 ± 11 67 ± 8 52 ± 9*,† 0.025 Peak ex 153 ± 20 142 ± 13 138 ± 20 0.238 Mean arterial pressure (mmHg) Rest 88 ± 7 90 ± 10 81 ± 11 0.317 Peak ex 113 ± 14 117 ± 15 109 ± 7 0.580 Ea (mmHg/mL) Rest 1.9 ± 0.4 2.6 ± 0.6* 1.6 ± 0.4 0.003 Peak ex 2.1 ± 0.5 2.9 ± 0.3* 2.1 ± 0.3 <0.001 LVEDV (mL/m2) Rest 122 ± 18 99 ± 23* 133 ± 11† 0.007 Peak ex 124 ± 23 104 ± 26 134 ± 15 0.058 RVEDV (mL/m2) Rest 124 ± 22 81 ± 18* 125 ± 19† <0.001 Peak ex 119 ± 23 77 ± 20* 122 ± 22† <0.001 LVESV (mL/m2) Rest 59 ± 9† 52 ± 15 63 ± 17 0.272 Peak ex 44 ± 9 51 ± 22 58 ± 17 0.298 RVESV (mL/m2) Rest 59 ± 10 34 ± 10* 58 ± 2† <0.001 Peak ex 40 ± 9 26 ± 10 51 ± 17† 0.003 LVSV (mL/m2) Rest 63 ± 11 47 ± 10* 70 ± 9† 0.001 Peak ex 81 ± 15 53 ± 9* 76 ± 8† <0.001 RVSV (mL/m2) Rest 64 ± 11 47 ± 9* 67 ± 8† 0.001 Peak ex 79 ± 15 51 ± 10* 71 ± 7† <0.001 LVEF (%) Rest 51.1 ± 2.8 47.7 ± 5.5 53.0 ± 9.2 0.215 Peak ex 64.8 ± 3.7 52.8 ± 9.5* 57.2 ± 8.9 0.007 RVEF (%) Rest 51.9 ± 2.1 58.4 ± 6.6* 53.9 ± 3.3 0.017 Peak ex 66.3 ± 4.0 66.6 ± 5.9 59.0 ± 7.5 0.047 CI (L/min/m2) Rest 4.3 ± 1.4 3.1 ± 0.6 3.5 ± 0.5 0.070 Peak ex 12.3 ± 3.4 7.4 ± 1.6* 10.1 ± 1.1 0.001 LAVmax (mL/m2) Rest 52 ± 8 46 ± 15 57 ± 11 0.219 Peak ex 56 ± 18 53 ± 15 61 ± 20 0.745 RAVmax (mL/m2) Rest 78 ± 9 63 ± 26 95 ± 18† 0.020 Peak ex 68 ± 22 66 ± 29 76 ± 22 0.755 LAVmin (mL/m2) Rest 27 ± 5 24 ± 8 36 ± 10† 0.020 Peak ex 25 ± 8 31 ± 13 36 ± 11 0.202 RAVmin (mL/m2) Rest 42 ± 8 34 ± 15 59 ± 10*,† 0.003 Peak ex 26 ± 9 34 ± 18 40 ± 13 0.188 EA-healthy (n = 10) DCM (n = 9) EA-fibrosis (n = 5) P-value Heart rate (b.p.m.) Rest 66 ± 11 67 ± 8 52 ± 9*,† 0.025 Peak ex 153 ± 20 142 ± 13 138 ± 20 0.238 Mean arterial pressure (mmHg) Rest 88 ± 7 90 ± 10 81 ± 11 0.317 Peak ex 113 ± 14 117 ± 15 109 ± 7 0.580 Ea (mmHg/mL) Rest 1.9 ± 0.4 2.6 ± 0.6* 1.6 ± 0.4 0.003 Peak ex 2.1 ± 0.5 2.9 ± 0.3* 2.1 ± 0.3 <0.001 LVEDV (mL/m2) Rest 122 ± 18 99 ± 23* 133 ± 11† 0.007 Peak ex 124 ± 23 104 ± 26 134 ± 15 0.058 RVEDV (mL/m2) Rest 124 ± 22 81 ± 18* 125 ± 19† <0.001 Peak ex 119 ± 23 77 ± 20* 122 ± 22† <0.001 LVESV (mL/m2) Rest 59 ± 9† 52 ± 15 63 ± 17 0.272 Peak ex 44 ± 9 51 ± 22 58 ± 17 0.298 RVESV (mL/m2) Rest 59 ± 10 34 ± 10* 58 ± 2† <0.001 Peak ex 40 ± 9 26 ± 10 51 ± 17† 0.003 LVSV (mL/m2) Rest 63 ± 11 47 ± 10* 70 ± 9† 0.001 Peak ex 81 ± 15 53 ± 9* 76 ± 8† <0.001 RVSV (mL/m2) Rest 64 ± 11 47 ± 9* 67 ± 8† 0.001 Peak ex 79 ± 15 51 ± 10* 71 ± 7† <0.001 LVEF (%) Rest 51.1 ± 2.8 47.7 ± 5.5 53.0 ± 9.2 0.215 Peak ex 64.8 ± 3.7 52.8 ± 9.5* 57.2 ± 8.9 0.007 RVEF (%) Rest 51.9 ± 2.1 58.4 ± 6.6* 53.9 ± 3.3 0.017 Peak ex 66.3 ± 4.0 66.6 ± 5.9 59.0 ± 7.5 0.047 CI (L/min/m2) Rest 4.3 ± 1.4 3.1 ± 0.6 3.5 ± 0.5 0.070 Peak ex 12.3 ± 3.4 7.4 ± 1.6* 10.1 ± 1.1 0.001 LAVmax (mL/m2) Rest 52 ± 8 46 ± 15 57 ± 11 0.219 Peak ex 56 ± 18 53 ± 15 61 ± 20 0.745 RAVmax (mL/m2) Rest 78 ± 9 63 ± 26 95 ± 18† 0.020 Peak ex 68 ± 22 66 ± 29 76 ± 22 0.755 LAVmin (mL/m2) Rest 27 ± 5 24 ± 8 36 ± 10† 0.020 Peak ex 25 ± 8 31 ± 13 36 ± 11 0.202 RAVmin (mL/m2) Rest 42 ± 8 34 ± 15 59 ± 10*,† 0.003 Peak ex 26 ± 9 34 ± 18 40 ± 13 0.188 LAVmax, left atrial maximal volume; LAVmin, left atrial minimal volume; RAVmax, right atrial maximal volume; RAVmin, right atrial minimal volume; RVEDV, right ventricular end-diastolic volume. * P < 0.05 for difference vs. EA-healthy. †P < 0.05 for difference vs. DCM. Table 2 Cardiac haemodynamics EA-healthy (n = 10) DCM (n = 9) EA-fibrosis (n = 5) P-value Heart rate (b.p.m.) Rest 66 ± 11 67 ± 8 52 ± 9*,† 0.025 Peak ex 153 ± 20 142 ± 13 138 ± 20 0.238 Mean arterial pressure (mmHg) Rest 88 ± 7 90 ± 10 81 ± 11 0.317 Peak ex 113 ± 14 117 ± 15 109 ± 7 0.580 Ea (mmHg/mL) Rest 1.9 ± 0.4 2.6 ± 0.6* 1.6 ± 0.4 0.003 Peak ex 2.1 ± 0.5 2.9 ± 0.3* 2.1 ± 0.3 <0.001 LVEDV (mL/m2) Rest 122 ± 18 99 ± 23* 133 ± 11† 0.007 Peak ex 124 ± 23 104 ± 26 134 ± 15 0.058 RVEDV (mL/m2) Rest 124 ± 22 81 ± 18* 125 ± 19† <0.001 Peak ex 119 ± 23 77 ± 20* 122 ± 22† <0.001 LVESV (mL/m2) Rest 59 ± 9† 52 ± 15 63 ± 17 0.272 Peak ex 44 ± 9 51 ± 22 58 ± 17 0.298 RVESV (mL/m2) Rest 59 ± 10 34 ± 10* 58 ± 2† <0.001 Peak ex 40 ± 9 26 ± 10 51 ± 17† 0.003 LVSV (mL/m2) Rest 63 ± 11 47 ± 10* 70 ± 9† 0.001 Peak ex 81 ± 15 53 ± 9* 76 ± 8† <0.001 RVSV (mL/m2) Rest 64 ± 11 47 ± 9* 67 ± 8† 0.001 Peak ex 79 ± 15 51 ± 10* 71 ± 7† <0.001 LVEF (%) Rest 51.1 ± 2.8 47.7 ± 5.5 53.0 ± 9.2 0.215 Peak ex 64.8 ± 3.7 52.8 ± 9.5* 57.2 ± 8.9 0.007 RVEF (%) Rest 51.9 ± 2.1 58.4 ± 6.6* 53.9 ± 3.3 0.017 Peak ex 66.3 ± 4.0 66.6 ± 5.9 59.0 ± 7.5 0.047 CI (L/min/m2) Rest 4.3 ± 1.4 3.1 ± 0.6 3.5 ± 0.5 0.070 Peak ex 12.3 ± 3.4 7.4 ± 1.6* 10.1 ± 1.1 0.001 LAVmax (mL/m2) Rest 52 ± 8 46 ± 15 57 ± 11 0.219 Peak ex 56 ± 18 53 ± 15 61 ± 20 0.745 RAVmax (mL/m2) Rest 78 ± 9 63 ± 26 95 ± 18† 0.020 Peak ex 68 ± 22 66 ± 29 76 ± 22 0.755 LAVmin (mL/m2) Rest 27 ± 5 24 ± 8 36 ± 10† 0.020 Peak ex 25 ± 8 31 ± 13 36 ± 11 0.202 RAVmin (mL/m2) Rest 42 ± 8 34 ± 15 59 ± 10*,† 0.003 Peak ex 26 ± 9 34 ± 18 40 ± 13 0.188 EA-healthy (n = 10) DCM (n = 9) EA-fibrosis (n = 5) P-value Heart rate (b.p.m.) Rest 66 ± 11 67 ± 8 52 ± 9*,† 0.025 Peak ex 153 ± 20 142 ± 13 138 ± 20 0.238 Mean arterial pressure (mmHg) Rest 88 ± 7 90 ± 10 81 ± 11 0.317 Peak ex 113 ± 14 117 ± 15 109 ± 7 0.580 Ea (mmHg/mL) Rest 1.9 ± 0.4 2.6 ± 0.6* 1.6 ± 0.4 0.003 Peak ex 2.1 ± 0.5 2.9 ± 0.3* 2.1 ± 0.3 <0.001 LVEDV (mL/m2) Rest 122 ± 18 99 ± 23* 133 ± 11† 0.007 Peak ex 124 ± 23 104 ± 26 134 ± 15 0.058 RVEDV (mL/m2) Rest 124 ± 22 81 ± 18* 125 ± 19† <0.001 Peak ex 119 ± 23 77 ± 20* 122 ± 22† <0.001 LVESV (mL/m2) Rest 59 ± 9† 52 ± 15 63 ± 17 0.272 Peak ex 44 ± 9 51 ± 22 58 ± 17 0.298 RVESV (mL/m2) Rest 59 ± 10 34 ± 10* 58 ± 2† <0.001 Peak ex 40 ± 9 26 ± 10 51 ± 17† 0.003 LVSV (mL/m2) Rest 63 ± 11 47 ± 10* 70 ± 9† 0.001 Peak ex 81 ± 15 53 ± 9* 76 ± 8† <0.001 RVSV (mL/m2) Rest 64 ± 11 47 ± 9* 67 ± 8† 0.001 Peak ex 79 ± 15 51 ± 10* 71 ± 7† <0.001 LVEF (%) Rest 51.1 ± 2.8 47.7 ± 5.5 53.0 ± 9.2 0.215 Peak ex 64.8 ± 3.7 52.8 ± 9.5* 57.2 ± 8.9 0.007 RVEF (%) Rest 51.9 ± 2.1 58.4 ± 6.6* 53.9 ± 3.3 0.017 Peak ex 66.3 ± 4.0 66.6 ± 5.9 59.0 ± 7.5 0.047 CI (L/min/m2) Rest 4.3 ± 1.4 3.1 ± 0.6 3.5 ± 0.5 0.070 Peak ex 12.3 ± 3.4 7.4 ± 1.6* 10.1 ± 1.1 0.001 LAVmax (mL/m2) Rest 52 ± 8 46 ± 15 57 ± 11 0.219 Peak ex 56 ± 18 53 ± 15 61 ± 20 0.745 RAVmax (mL/m2) Rest 78 ± 9 63 ± 26 95 ± 18† 0.020 Peak ex 68 ± 22 66 ± 29 76 ± 22 0.755 LAVmin (mL/m2) Rest 27 ± 5 24 ± 8 36 ± 10† 0.020 Peak ex 25 ± 8 31 ± 13 36 ± 11 0.202 RAVmin (mL/m2) Rest 42 ± 8 34 ± 15 59 ± 10*,† 0.003 Peak ex 26 ± 9 34 ± 18 40 ± 13 0.188 LAVmax, left atrial maximal volume; LAVmin, left atrial minimal volume; RAVmax, right atrial maximal volume; RAVmin, right atrial minimal volume; RVEDV, right ventricular end-diastolic volume. * P < 0.05 for difference vs. EA-healthy. †P < 0.05 for difference vs. DCM. At rest, LVEF was similar in the different groups. In the DCM and EA-fibrosis groups, less augmentation of LVEF (5 ± 6% vs. 4 ± 3% vs. 14 ± 3%; P = 0.001; Figure 2) and contractility [LVESPVR ratio; 1.4 (1.3–1.6) vs. 1.5 (1.3–1.6) vs. 1.8 (1.7–2.7); P < 0.001] was observed during exercise than EA-healthy. This was due to an attenuated reduction in LVESVi (Figure 3), whereas the response of LVEDVi was similar in all groups. Exercise-induced changes in Ea were similar between groups (+14 ± 23% vs. +13 ± 21% vs. 27 ± 8%; P = 0.422), whereas the change in LVESPVR was greater in EA-healthy compared with DCM and EA-fibrosis (+105 ± 51% vs. +38 ± 17% vs. +53 ± 18%). Figure 2 View largeDownload slide LVEF during exercise. All EA-healthy subjects demonstrate an increase in ΔLVEF during exercise as opposed to DCM patients and EA-fibrosis subjects in whom a heterogeneous response is observed (A). Asterisk indicates P-value <0.05 for the difference in ΔLVEF vs. EA-healthy. Error bars denote standard deviation. (B) End-systolic images at rest and at peak-exercise in a healthy EA, a patient with mild DCM and an EA with fibrosis. In the EA-healthy, LV and RV function augment with exercise. In contrast, LV function fails to augment in the DCM patient and the EA with fibrosis (right image). Figure 2 View largeDownload slide LVEF during exercise. All EA-healthy subjects demonstrate an increase in ΔLVEF during exercise as opposed to DCM patients and EA-fibrosis subjects in whom a heterogeneous response is observed (A). Asterisk indicates P-value <0.05 for the difference in ΔLVEF vs. EA-healthy. Error bars denote standard deviation. (B) End-systolic images at rest and at peak-exercise in a healthy EA, a patient with mild DCM and an EA with fibrosis. In the EA-healthy, LV and RV function augment with exercise. In contrast, LV function fails to augment in the DCM patient and the EA with fibrosis (right image). Figure 3 View largeDownload slide Changes in biventricular end-systolic volume during exercise. Changes in LVESVi (A) and RV ESVi (B) during incremental exercise are shown for EA-healthy (green), EA-fibrosis (black) and DCM patients (red). P-values are shown for the interaction between group and exercise-intensity. Error bars denote standard error of the mean. Figure 3 View largeDownload slide Changes in biventricular end-systolic volume during exercise. Changes in LVESVi (A) and RV ESVi (B) during incremental exercise are shown for EA-healthy (green), EA-fibrosis (black) and DCM patients (red). P-values are shown for the interaction between group and exercise-intensity. Error bars denote standard error of the mean. In a sub-analysis comparing EA-fibrosis and EA-healthy (n = 5), matched for exercise capacity (VO2peak 68.4 ± 4.2 vs. 64.0 ± 5.9 mL/kg/min; P = 0.210), the difference in LVEF reserve remained significant (ΔLVEF 4 ± 3% vs. 14 ± 2%; P < 0.001). DCM patients had higher resting RVEF compared with EA-healthy and EA-fibrosis subjects, the latter groups having similar values. As compared with EA-healthy, the exercise-induced increase in RVEF was diminished in both DCM and EA-fibrosis (ΔRVEF 8 ± 6% vs. 5 ± 5% vs. 14 ± 4%; P = 0.006) due to an attenuated reduction in RVESVi (P = 0.001 for interaction). Similar to LV functional measures, neither DCM nor EA-fibrosis demonstrated an increase in LAEF or RAEF during exercise, whereas LAEF and RAEF increased in EA-healthy (Figure 4). Figure 4 View largeDownload slide Changes in ventricular and atrial function during exercise. Changes in LVEF (A), RVEF (B) and LAEF (C) and RAEF (D) during incremental exercise are shown for EA-healthy (green), EA-fibrosis (black) and DCM patients (red). P-values are shown for the interaction between group and exercise-intensity. Error bars denote standard error of the mean. Figure 4 View largeDownload slide Changes in ventricular and atrial function during exercise. Changes in LVEF (A), RVEF (B) and LAEF (C) and RAEF (D) during incremental exercise are shown for EA-healthy (green), EA-fibrosis (black) and DCM patients (red). P-values are shown for the interaction between group and exercise-intensity. Error bars denote standard error of the mean. An example of exercise CMR comparing LV function in a DCM patient with EA-healthy and EA-fibrosis subjects is provided in Figure 2 and Supplementary data online, Video S1. Diagnostic accuracy to differentiate EA from DCM patients Receiver-operator characteristic curves demonstrated that cut-off values of 11.2% for the increase in LVEF from rest to peak exercise (AUC = 0.92, P < 0.001) had a sensitivity of 93% and specificity of 90% to differentiate EA-healthy from DCM and EA-fibrosis, whereas the LVESPVR ratio of 1.8 (AUC = 0.94; P = 0.003) had a sensitivity of 83% and a specificity of 100% patients; Figure 5). In contrast, resting LVEF was not predictive (AUC = 0.56, P = 0.598). Similarly, VO2peak was unable to accurately separate the cohorts (AUC = 0.68, P = 0.128). The AUC value for ΔLVEF was statistically different from that of resting LVEF (P = 0.013) and similar to that of the LVESPVR ratio (P = 0.879). Figure 5 View largeDownload slide Differentiation between physiological and pathological left ventricular remodelling. Receiver operating curves for the ability of resting and exercise measurements of LVEF and ΔLVEF, peak oxygen consumption (VO2peak), and contractility (end-systolic pressure–volume ratio) to differentiate healthy EAs with physiological remodelling from DCM patients and EAs with fibrosis. Figure 5 View largeDownload slide Differentiation between physiological and pathological left ventricular remodelling. Receiver operating curves for the ability of resting and exercise measurements of LVEF and ΔLVEF, peak oxygen consumption (VO2peak), and contractility (end-systolic pressure–volume ratio) to differentiate healthy EAs with physiological remodelling from DCM patients and EAs with fibrosis. Discussion The main goal of this study was to determine whether exercise evaluation of cardiac reserve enables differentiation between physiological adaptation to endurance exercise and pathological LV remodelling. Whereas LV systolic function was similar between athletes and DCM at rest, exercise imaging reliably distinguished between healthy athletes and those with pathology (DCM and athletes with LV fibrosis). Importantly, by comparing healthy athletes to elite EA with manifest LV fibrosis, we demonstrated that exercise capacity alone does not exclude significant LV damage. Therefore, evaluation of contractile reserve may be a useful tool, which can be applied in both athletic and non-athletic populations to separate health from disease. Although multiple studies reported profound increases in LV and RV mass and volumes due to intensive endurance training, significant variability exists in the extent of cardiac remodelling. In a study on Tour de France cyclists, Abergel et al.1 reported that about one-half of athletes had substantial LV enlargement and 12% had an LVEF <52%, highlighting the diagnostic overlap between health and DCM. Accurate differentiation between these entities is of paramount importance. Incorrectly labelling an athlete with pathology could lead to inappropriate exclusion from sport and, of even greater consequence, failing to identify early pathology will mean that treatments with proven prognostic benefit (e.g. ACE inhibitors) may be delayed, and the patient may be exposed to an increased risk of cardiomyopathy and arrhythmic sudden cardiac death.2,18 Current diagnostic tests rely on resting assessments and are unproven. Similarly, temporal cessation of sports activity (‘detraining’) incurs significant periods of uncertainty for the athlete and has not been shown to assist in decision-making. Given that athletic ventricular enlargement has been observed to persist for many years,19 it is unlikely that short-term detraining could reliably distinguish athletic remodelling from pathology. Our current results demonstrate that evaluation during exercise facilitates the differentiation between athlete’s heart and pathological LV adaptation with considerable accuracy. In both groups with LV pathology, there was impaired contractile reserve characterized by failure to decrease end-systolic volumes and increase LVEF.20 In contrast, healthy EAs demonstrated a consistent reduction in LV end-systolic volumes and a significant increase in LVEF, irrespective of resting LVEF. Although criticism is often levelled at the load-dependent volume measures and EF, we found no advantage in combining these measures with invasively derived pressures. The most direct assessment of contractility would be to adjust volumes for pressure, and thus, we assessed the LV end-systolic pressure–volume relationship as a surrogate of contractility. DCM and EA-fibrosis subjects had similar changes in arterial elastance but reduced contractile reserve. Taken together, these findings suggest exercise-induced uncoupling between the LV and the arterial system, as evidenced by impaired augmentation of LVEF, a non-invasive surrogate of ventricular-arterial coupling.21 Importantly, we demonstrated that purely non-invasive volumetric measures proved as accurate in separating health from disease. This provides obvious advantages for the future utility of these techniques. Future work will have to determine whether exercise echocardiography is as reliable as our CMR evaluation, although image quality and through-plane movement during exercise provide limitations in feasibility and reproducibility.8 An important finding of this study is that maximal peak oxygen consumption was paradoxically higher in athletes with LV damage compared with healthy EA, while peak CI was similar in both groups. In combination, these findings suggest that peripheral factors play an important role in the preservation of cardiopulmonary exercise performance despite a reduction in LV contractile reserve. This finding is in keeping with previous data in hypertrophic cardiomyopathy in which exercise testing facilitated the diagnosis in only one-fifth of subjects.6 The EA with fibrosis in the current study were high-level competitive athletes as evidenced by their excellent exercise capacity. Hence, all of them had a VO2peak >60 mL/kg/min. Nevertheless, they all presented with ventricular arrhythmias, either at initial workup or during follow-up, as previously reported, indicating that superb exercise capacity per se does not imply a good prognosis.4 Another important finding of this study is that the functional abnormalities during exercise were not restricted to the LV, but also involved the RV and both atria. Previous studies using radionuclide ventriculography or invasive techniques have assessed either the LV16 or the RV22 response to exercise in severe DCM. In addition, several studies used pharmacological stress echocardiography or stress CMR to assess LV or biventricular contractile reserve.23–25 To date, however, no studies have evaluated the response of both ventricles and atria to incremental exercise simultaneously. In keeping with previous data, we found that the increase in RVEF was diminished in the DCM and particularly the EA-fibrosis group compared with healthy EAs. Furthermore, changes in bi-atrial function during exercise mirrored the changes observed in the ventricles, as evidenced by attenuated increases in LA and RA emptying function. This is in line with previous observations revealing that the degree of left atrial dysfunction at rest is greater than expected from the degree of atrial dilatation, thereby suggesting myopathy of the left atrium.26 Moreover, there is data to suggest that left atrial myopathy may even precede LV myopathy and can be used for early detection and risk stratification in some cases.27,28 In this context, the normal functional response of both atria to exercise in EAs in the current study provides additional reassurance that the low resting LVEF in EAs is not due to underlying myocardial pathology, but rather represents physiological remodelling. The findings of this study have clinical implications for subjects at risk for the development of DCM, e.g. asymptomatic family members of patients with severe DCM or prior myocarditis. Due to the design of this study, comparing contractile reserve in EAs and patients with only mild DCM, the degree of resting LV impairment in the DCM group was substantially less profound than in previous studies evaluating contractile reserve in DCM.14,22,23 As a result, exercise capacity was preserved, NTproBNP was only mildly elevated and the diagnosis of mild DCM was mainly based on a mildly reduced resting LVEF and a high index of suspicion due to disease in other family members. As such, these subjects were well suited for comparison with the EAs because the differentiation from Athlete’s Heart would be clinically obvious in patients with severe DCM. Until present, there had been no studies evaluating LV contractile reserve in this subset of DCM patients with mild phenotype despite the recommendation to perform exercise imaging as part of the diagnostic workup.7 We observed that LV functional reserve in the DCM patients was heterogeneous and independent of resting LVEF. This is consistent with previous data in patients with severe DCM,14 which suggests that LV contractile reserve may improve clinical risk stratification and guide therapeutic management of these subjects in the early course of the disease. Further studies are required to verify whether those subjects with higher contractile reserve have better outcome as compared with those in whom cardiac function deteriorates during exercise. Limitations Firstly, the comprehensive measurements undertaken in this proof-of-concept study and the unique characteristics of the cohorts limited the sample size. The small sample size may have increased the probability of Type II statistical errors while multiple comparisons increase the likelihood of Type I errors. Nevertheless, to date this is the only study to compare exercise measures in EAs and DCM patients. It would have been valuable to include a control group of athletes with established DCM. However, given the low prevalence of DCM amongst athletes, we included a cohort of elite EA with underlying LV damage associated with ventricular arrhythmias. This is the largest cohort to date to assess cardiac function and exercise metrics. The established accuracy of exercise CMR measures enabled us to evaluate meaningful haemodynamic differences even within this modest-sized cohort. The fact that highly significant differences were apparent reinforces the accuracy of ex-CMR and the extent of the physiological differences. Further larger and prospective studies are required to validate the cut-off values reported in this study. LVESPVR is not equal to end-systolic elastance. Calculation of V0 typically requires the use of invasively derived pressure–volume loops, which was considered far too invasive for the scope of this study. Nevertheless, LVESPVR is considered a valid approximation of end-systolic elastance.11,29 We cannot exclude that V0 changes during exercise. However, previous studies reported that V0 remains unaltered during exercise or changes in loading conditions.30,31 As expected, the majority of DCM patients had received treatment including beta-blockers or calcium blockers. This introduces confounders in comparisons with the EAs and control subjects. However, beta-blocker and calcium channel blocking medications were withheld for at least 24 h prior to exercise testing; a period sufficient to exclude any persisting pharmacodynamic effect. Finally, although the cardiac response to exercise in the EAs was similar to previous observations in healthy athletes,9 longer follow-up is necessary to assess whether the mildly reduced resting LV function is associated with clinical events, e.g. arrhythmias, in the long term. Similarly, larger studies are required to verify whether assessment of contractile reserve provides additional value for risk stratification in DCM patients with mildly reduced LVEF. Conclusions Neither exercise capacity nor resting measures of cardiac function are helpful in differentiating healthy athletes with borderline LVEF from those with myocardial impairment or fibrosis. On the other hand, augmentation of biventricular and atrial function during exercise are important tools that can accurately differentiate between physiological and pathological LV remodelling. Supplementary data Supplementary data are available at European Heart Journal - Cardiovascular Imaging online. Funding This study was funded by a grant from the Fund for Scientific Research Flanders (FWO), Belgium. Andre La Gerche was funded by a Career Development Fellowship from the National Health and Medical Research Council and National Heart Foundation of Australia (NHMRC and NHF). Frédéric Schnell was funded by a grant of the French Federation of Cardiology. Conflict of interest: None declared. References 1 Abergel E , Chatellier G , Hagege AA , Oblak A , Linhart A , Ducardonnet A et al. Serial left ventricular adaptations in world-class professional cyclists: implications for disease screening and follow-up . J Am Coll Cardiol 2004 ; 44 : 144 – 9 . Google Scholar CrossRef Search ADS PubMed 2 Maron BJ , Doerer JJ , Haas TS , Tierney DM , Mueller FO. Sudden deaths in young competitive athletes: analysis of 1866 deaths in the United States, 1980-2006 . Circulation 2009 ; 119 : 1085 – 92 . Google Scholar CrossRef Search ADS PubMed 3 Heidbuchel H , Hoogsteen J , Fagard R , Vanhees L , Ector H , Willems R et al. High prevalence of right ventricular involvement in endurance athletes with ventricular arrhythmias. Role of an electrophysiologic study in risk stratification . Eur Heart J 2003 ; 24 : 1473 – 80 . Google Scholar CrossRef Search ADS PubMed 4 Schnell F , Claessen G , La Gerche A , Bogaert J , Lentz PA , Claus P et al. Subepicardial delayed gadolinium enhancement in asymptomatic athletes: let sleeping dogs lie? Br J Sports Med 2016 ; 50 : 111 – 7 . Google Scholar CrossRef Search ADS PubMed 5 Trivax JE , McCullough PA. Phidippides cardiomyopathy: a review and case illustration . Clin Cardiol 2012 ; 35 : 69 – 73 . Google Scholar CrossRef Search ADS PubMed 6 Sheikh N , Papadakis M , Schnell F , Panoulas V , Malhotra A , Wilson M et al. Clinical profile of athletes with hypertrophic cardiomyopathy . Circ Cardiovasc Imaging 2015 ; 8 : e003454. Google Scholar CrossRef Search ADS PubMed 7 Galderisi M , Cardim N , D'Andrea A , Bruder O , Cosyns B , Davin L et al. The multi-modality cardiac imaging approach to the Athlete's heart: an expert consensus of the European Association of Cardiovascular Imaging . Eur Heart J Cardiovasc Imaging 2015 ; 16 : 353. Google Scholar CrossRef Search ADS PubMed 8 Claessen G , La Gerche A , Voigt JU , Dymarkowski S , Schnell F , Petit T et al. Accuracy of echocardiography to evaluate pulmonary vascular and RV function during exercise . JACC Cardiovasc Imaging 2016 ; 9 : 532 – 43 . Google Scholar CrossRef Search ADS PubMed 9 La Gerche A , Claessen G , Dymarkowski S , Voigt JU , De Buck F , Vanhees L et al. Exercise-induced right ventricular dysfunction is associated with ventricular arrhythmias in endurance athletes . Eur Heart J 2015 ; 36 : 1998 – 2010 . Google Scholar CrossRef Search ADS PubMed 10 La Gerche A , Claessen G , Van de Bruaene A , Pattyn N , Van Cleemput J , Gewillig M et al. Cardiac MRI: a new gold standard for ventricular volume quantification during high-intensity exercise . Circ Cardiovasc Imaging 2013 ; 6 : 329 – 38 . Google Scholar CrossRef Search ADS PubMed 11 Kelly RP , Ting CT , Yang TM , Liu CP , Maughan WL , Chang MS et al. Effective arterial elastance as index of arterial vascular load in humans . Circulation 1992 ; 86 : 513 – 21 . Google Scholar CrossRef Search ADS PubMed 12 Pellicori P , Zhang J , Lukaschuk E , Joseph AC , Bourantas CV , Loh H et al. Left atrial function measured by cardiac magnetic resonance imaging in patients with heart failure: clinical associations and prognostic value . Eur Heart J 2015 ; 36 : 733 – 42 . Google Scholar CrossRef Search ADS PubMed 13 Hoit BD. Left atrial size and function: role in prognosis . J Am Coll Cardiol 2014 ; 63 : 493 – 505 . Google Scholar CrossRef Search ADS PubMed 14 Kirlin PC , Das S , Zijnen P , Wijns W , Domenicucci S , Roelandt J et al. The exercise response in idiopathic dilated cardiomyopathy . Clin Cardiol 1984 ; 7 : 205 – 10 . Google Scholar CrossRef Search ADS PubMed 15 Konishi T , Koyama T , Aoki T , Yada T , Futagami Y , Sekioka K et al. Assessments of left ventricular function during exercise in patients with dilated cardiomyopathy: comparison with ischemic cardiomyopathy . J Cardiol 1989 ; 19 : 797 – 804 . Google Scholar PubMed 16 Metz CE. Basic principles of ROC analysis . Semin Nucl Med 1978 ; 8 : 283 – 98 . Google Scholar CrossRef Search ADS PubMed 17 DeLong ER , DeLong DM , Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach . Biometrics 1988 ; 44 : 837 – 45 . Google Scholar CrossRef Search ADS PubMed 18 Corrado D , Basso C , Rizzoli G , Schiavon M , Thiene G. Does sports activity enhance the risk of sudden death in adolescents and young adults? J Am Coll Cardiol 2003 ; 42 : 1959 – 63 . Google Scholar CrossRef Search ADS PubMed 19 Pelliccia A , Maron BJ , De Luca R , Di Paolo FM , Spataro A , Culasso F. Remodeling of left ventricular hypertrophy in elite athletes after long-term deconditioning . Circulation 2002 ; 105 : 944 – 9 . Google Scholar CrossRef Search ADS PubMed 20 Haddad F , Vrtovec B , Ashley EA , Deschamps A , Haddad H , Denault AY. The concept of ventricular reserve in heart failure and pulmonary hypertension: an old metric that brings us one step closer in our quest for prediction . Curr Opin Cardiol 2011 ; 26 : 123 – 31 . Google Scholar CrossRef Search ADS PubMed 21 Cohen-Solal A , Faraggi M , Czitrom D , Le Guludec D , Delahaye N , Gourgon R. Left ventricular-arterial system coupling at peak exercise in dilated nonischemic cardiomyopathy . Chest 1998 ; 113 : 870 – 7 . Google Scholar CrossRef Search ADS PubMed 22 Plehn G , Vormbrock J , Perings S , Plehn A , Meissner A , Butz T et al. Comparison of right ventricular functional response to exercise in hypertrophic versus idiopathic dilated cardiomyopathy . Am J Cardiol 2010 ; 105 : 116 – 21 . Google Scholar CrossRef Search ADS PubMed 23 Matsumoto K , Tanaka H , Onishi A , Motoji Y , Tatsumi K , Sawa T et al. Bi-ventricular contractile reserve offers an incremental prognostic value for patients with dilated cardiomyopathy . Eur Heart J Cardiovasc Imaging 2015 ; 16 : 1213 – 23 . Google Scholar CrossRef Search ADS PubMed 24 Pratali L , Picano E , Otasevic P , Vigna C , Palinkas A , Cortigiani L et al. Prognostic significance of the dobutamine echocardiography test in idiopathic dilated cardiomyopathy . Am J Cardiol 2001 ; 88 : 1374 – 8 . Google Scholar CrossRef Search ADS PubMed 25 Naqvi TZ , Goel RK , Forrester JS , Siegel RJ. Myocardial contractile reserve on dobutamine echocardiography predicts late spontaneous improvement in cardiac function in patients with recent onset idiopathic dilated cardiomyopathy . J Am Coll Cardiol 1999 ; 34 : 1537 – 44 . Google Scholar CrossRef Search ADS PubMed 26 Triposkiadis F , Pitsavos C , Boudoulas H , Trikas A , Toutouzas P. Left atrial myopathy in idiopathic dilated cardiomyopathy . Am Heart J 1994 ; 128 : 308 – 15 . Google Scholar CrossRef Search ADS PubMed 27 Graber HL , Unverferth DV , Baker PB , Ryan JM , Baba N , Wooley CF. Evolution of a hereditary cardiac conduction and muscle disorder: a study involving a family with six generations affected . Circulation 1986 ; 74 : 21 – 35 . Google Scholar CrossRef Search ADS PubMed 28 Otway R , Vandenberg JI , Guo G , Varghese A , Castro ML , Liu J et al. Stretch-sensitive KCNQ1 mutation A link between genetic and environmental factors in the pathogenesis of atrial fibrillation? J Am Coll Cardiol 2007 ; 49 : 578 – 86 . Google Scholar CrossRef Search ADS PubMed 29 Sagawa K , Suga H , Shoukas AA , Bakalar KM. End-systolic pressure/volume ratio: a new index of ventricular contractility . Am J Cardiol 1977 ; 40 : 748 – 53 . Google Scholar CrossRef Search ADS PubMed 30 Little WC , Cheng CP , Peterson T , Vinten-Johansen J. Response of the left ventricular end-systolic pressure-volume relation in conscious dogs to a wide range of contractile states . Circulation 1988 ; 78 : 736 – 45 . Google Scholar CrossRef Search ADS PubMed 31 Maughan WL , Sunagawa K , Burkhoff D , Sagawa K. Effect of arterial impedance changes on the end-systolic pressure-volume relation . Circ Res 1984 ; 54 : 595 – 602 . Google Scholar CrossRef Search ADS PubMed Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2018. For permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

European Heart Journal – Cardiovascular ImagingOxford University Press

Published: Mar 26, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off