Impact of Myocardial Fibrosis on Left Ventricular Function Evaluated by Feature-Tracking Myocardial Strain Cardiac Magnetic Resonance in Competitive Male Triathletes With Normal Ejection Fraction

Abstract

Background: To analyze the effect of myocardial fibrosis on left ventricular (LV) function evaluated by feature-tracking strain analysis by cine cardiac magnetic resonance (CMR) in competitive male triathletes with normal ejection fraction (EF).

Methods and Results: 78 asymptomatic male triathletes with >10 weekly training hours (43±11 years) and 28 male age-matched controls were studied by late gadolinium enhancement (LGE) and cine CMR. Global and segmental radial, longitudinal and circumferential strains were analyzed using feature-tracking cine CMR. Focal non-ischemic LGE was observed in 15 of 78 triathletes (19%, LGE+) with predominance in the basal inferolateral segments. LVEF was normal in LGE+ (62±6%) and LGE− triathletes (62±5%, P=0.958). In contrast, global radial strain was lower in LGE+ triathletes at 40±7% compared with LGE− triathletes (45±7%, P<0.05). Reduced segmental radial strain occurred either in LGE+ segments or in directly adjacent segments. Strain analysis revealed regional differences in controls, with the highest radial and longitudinal strain in the inferolateral segments, which were typically affected by fibrosis in LGE+ triathletes.

Conclusions: Reduced global and regional radial strain suggests a negative effect of myocardial fibrosis on LV function in LGE+ triathletes with normal EF. The observed regional differences in controls with the highest radial and longitudinal strain in the inferolateral segments may explain the typical occurrence of fibrosis in this myocardial region in triathletes.

Myocardial fibrosis detected by late gadolinium enhancement (LGE) cardiac magnetic resonance (CMR) occurs in a variable number of athletes, with studies reporting a prevalence between 0% and 50%.^1–7 The pattern of myocardial fibrosis in athletes suggests either an ischemic or non-ischemic genesis and several causes of non-ischemic fibrosis are currently discussed, including silent myocarditis, pulmonary artery pressure overload and repetitive myocardial microdamage.⁸ A recent work reported a series of 83 triathletes with myocardial fibrosis detected in 17% of the male triathletes on LGE CMR, but in none of the female triathletes.⁹ Non-ischemic LGE was observed in this study with predominant involvement of the inferolateral myocardium of the left ventricle (LV). In Anderson-Fabry disease it is suggested that increased wall stress and impaired resistance to physical stress are potential explanations for the typical inferolateral LGE involvement in such patients.¹⁰ However, the reasons for the noticeable involvement of this region in male triathletes remain to be elucidated.

Myocardial strain analysis using feature-tracking enables the analysis of global and segmental myocardial systolic contraction using standard cine CMR images.¹¹ It has shown that myocardial strain detects impaired systolic function in symptomatic patients with preserved ejection fraction (EF),¹² and that myocardial strain is superior to EF for predicting survival in patients with ischemic or non-ischemic cardiomyopathy.^13,14 We hypothesized that global and segmental myocardial strain would detect myocardial contraction anomalies in triathletes with myocardial fibrosis (LGE+) compared with triathletes without myocardial fibrosis (LGE−).

The purpose of the study was to analyze the effect of myocardial fibrosis on LV function evaluated by feature-tracking myocardial strain by cine CMR in competitive male triathletes with normal EF.

Methods

Triathletes and Controls

The institutional ethics committee approved the study and all subjects gave written informed consent. The authors had full control of the data and the material submitted for publication. Male triathletes were contacted through advertisement at local triathlon clubs. Inclusion criteria were weekly training ≥10 h and regular participation in triathlon competitions of various distances in the past 3 years. Triathletes reported their regular training hours per week and the number of active training years. A total of 79 competitive male triathletes (age: 43±11 years, range 18–66) and 28 male control subjects with a similar age distribution (40±10 years) were enrolled between April 2014 and December 2017 (Figure 1A). Control subjects were eligible if they exercised <3 h/week. Study exclusion criteria were contraindications for CMR or any systemic diseases. One male triathlete with known arterial hypertension was excluded from the study. None of the 78 triathletes and controls had any history of cardiovascular disease and reported no intake of any cardiac or illicit medications. All subjects underwent the CMR study before the exercise test, which was performed on the same day. All subjects were instructed to arrive rested without exercising or drinking alcohol in the preceding 72 h. Food and caffeine intake was restricted in the preceding 3 h. Blood samples were drawn immediately before the CMR from an antecubital vein while resting supine for 5 min to obtain hematocrit, creatine kinase, high-sensitivity troponin T (hs-TnT), and N-terminal pro-B-type natriuretic peptide (NT-proBNP).

Figure 1.

(A) Subject flow chart. One triathlete was excluded because of arterial hypertension. (B) Short- and long-axis LGE images of a LGE+ triathlete and LGE− triathlete. The LGE+ triathlete had focal non-ischemic mid-myocardial LGE, typically located in the basal inferior and inferolateral segments (B, Upper row). Feature-tracking myocardial strain analysis was performed on cine CMR images using dedicated software, which automatically propagated the endo- and epicardial contours throughout the cardiac cycle, generating global and segmental myocardial peak strain values (B, Lower row). The LGE+ triathlete had reduced global peak radial strain with 40% (C, Left), whereas the LGE− triathlete had normal global peak radial strain with 50% (C, Right). Segmental strain analysis showed reduced regional radial strain in the basal inferolateral LGE+ segments (D, Left, marked in red) and in the adjacent mid-inferolateral segments. CMR, cardiac magnetic resonance; LGE, late gadolinium enhancement.

CMR Protocol

CMR was performed using a 1.5-T Achieva scanner equipped with a 5-channel cardiac phased array receive coil (Philips Healthcare, Best, The Netherlands). The CMR protocol included standard steady-state free-precession cine CMR in the short axis for the LV and right ventricle (RV) volumetry and LV mass, and 3 long-axis cine CMR images in standard 2-, 3- and 4-chamber views with the following typical imaging parameters: acquired voxel size (AVS) 1.98×1.80×6 mm³, reconstructed voxel size (RVS) 1.36×1.36×6 mm³, gap 4 mm, 9–10 slices for full LV coverage, echo time=1.67 ms, time to repetition=3.34 ms, flip angle=60°, parallel acquisition technique=SENSE factor 2.0, 25 phases per RR interval. Ten minutes after bolus injection of 0.2 mmol/kg gadoter acid (Dotarem^®, Guerbet, Sulzbach, Germany) at a rate of 2.5 mL/s end-diastolic LGE images were acquired using end-diastolic phase-sensitive inversion recovery sequences in short-axis orientation covering the entire heart and in two-, three- and four-chamber views with the following typical imaging parameters: AVS 1.59×1.71×8 mm³, RVS 0.97×0.98×8 mm³, gap 2 mm, 9–10 short-axis slices, echo time=2.40 ms, time to repetition=5.50 ms, flip angle=15°. The optimal inversion delay was obtained from a Look-Locker experiment.

LV Function on Cine CMR and LGE Quantification on Contrast-Enhanced CMR

Two investigators independently and blindly analyzed all CMR images and the data are given as the mean of the 2 observers. CMR parameters were indexed to the calculated body surface area (BSA). LV volume, mass and LGE size were evaluated using cvi42 software (Circle Cardiovascular Imaging Inc, Calgary, Alberta, Canada) in standard fashion on the short-axis cine images.¹⁵ Myocardial fibrosis was quantified using a threshold method with a cutoff >5 standard deviations (SD) above remote normal myocardium.¹⁵ The extent of global and segmental fibrosis was evaluated and reported as %LV and g/m² or % per segment, respectively. The distribution and pattern of LGE was visually analyzed and reported using the 16-segment model of the American Heart Association.¹⁶

Global and Segmental Strain Analysis

Myocardial strain was analyzed on cine CMR images using segment feature-tracking software version 2.1.R.6108 (Medviso, Lund, Sweden) with known excellent interobserver reproducibility.¹⁷ This software analyzes myocardial strain by computing the interframe deformation fields using an endocardial tracking strategy based on non-rigid image registration.^18,19 Instead of myocardial boundary tracking only, the software uses the entire image content (i.e., blood pool, entire myocardium) during the optimization process.¹⁸ Global and segmental longitudinal and radial strain were measured on 3 long-axis cine CMRs using standard 2-, 3- and 4-chamber views, whereas global and segmental circumferential strains were measured on 3 short-axis cine CMRs at the apical, middle and basal LV sections.¹⁸ Endo- and epicardial contours were manually delineated on end-diastolic images and were then automatically propagated by the software throughout the cardiac cycle generating myocardial strain values.¹⁸ Global and segmental strain were measured in controls and triathletes. For correlation between segmental strain and segmental LGE size the individual strains were normalized to the strain of the controls, because of the observed regional differences in myocardial strain, using the following formula: (individual segmental strain of a triathlete–mean segmental strain of controls)/mean segmental strain of controls×100.

Exercise Test

A ramp-incremental cardiopulmonary exercise test was performed on an eddy current brake cycle ergometer (Ergoselect 100, Ergoline GmbH, Bitz, Germany) to determine the maximal power and peak oxygen uptake (V̇O_2peak), which was calculated as the average of the 3 highest measurements within the final minute.²⁰ The 12-lead ECG and heart rate were monitored continuously, and blood pressure was automatically measured every 2 min using an upper arm cuff. The ramp-incremental exercise test was preceded by a 2-min resting period, followed by cycling at a low workload (20 W for 3 min) until a steady state of gas exchange was attained. Depending on the individual’s training history, the load was continuously increased in constant steps by 20–40 W/min to bring the participant to the limit of tolerance within 10–12 min of exercise.

Statistical Analysis

Statistical analysis was performed using MedCalc for Windows, version 13.3.3.0 (MedCalc Software, Ostend, Belgium) and SAS for Windows, version 9.4 (SA Institute Inc., Cary, NC, USA). Continuous data are presented as mean±SD and categorical data are presented as absolute numbers and percentages. Bland-Altman analysis was used to determine agreement between the 2 investigators.²¹ The intraclass correlation coefficient was calculated to analyze the inter-rater reliability. Continuous data were compared using two-sided Student t-tests and categorical variables were compared using Chi-squared test or Fisher’s exact test as appropriate. A one-way ANOVA test was used to compare strain values among the controls, LGE+ and LGE− triathletes. If significant differences were detected, all pairs of groups were tested with one linear mixed model to control for cumulative type I error, which occurs with multiple testing. Statistical significance was defined as P<0.05.

Results

Demographics and CMR Characteristics of Triathletes and Controls

No differences were observed between male triathletes and male controls regarding age, weight, height, BMI and BSA (Table 1). Resting heart rate was lower and peak systolic blood pressure under exercise was higher in the triathletes compared with controls. V̇O_2peak and maximal power were higher in the triathletes compared with controls. CMR revealed larger LV mass indices as well as higher left and right ventricular volume indices in the triathletes compared with controls (Table 1).

Table 1. Differences Between Male Triathletes and Controls

	Male triathletes (n=78)	Male controls (n=28)	P value
Clinical parameters
Age, years	43±11	40±10	0.241
Weight, kg	78±9	80±9	0.263
Height, m	1.81±0.07	1.81±0.09	0.689
BMI, kg/m2	23.7±2.3	24.5±2.2	0.072
BSA, m2	1.98±0.13	2.00±0.15	0.527
Exercise parameters
Systolic BP at rest, mmHg	125±15	122±14	0.531
Diastolic BP at rest, mmHg	83±10	85±11	0.511
Peak systolic BP, mmHg	199±28	181±34	<0.05
Peak diastolic BP, mmHg	91±21	82±21	0.087
HR at rest, beats/min	67±13	76±13	<0.01
Peak HR, beats/min	169±13	173±15	0.255
ΔHR rest/peak, beats/min	103±16	99±17	0.287
V̇O2peak, mL/kg/min	54±9	39±6	<0.0001
Maximal power, W	407±108	242±43	<0.0001
Blood parameters
Hematocrit, %	0.44±0.03	0.43±0.02	0.398
hs-TNT, pg/mL	7±8	5±3	0.323
NT-proBNP, pg/mL	41±60	34±23	0.579
CMR parameters
HR on CMR, beats/min	53±9	65±11	<0.0001
LVEF, %	62±5	61±8	0.519
Cardiac index, L/min/m2	3.3±0.7	3.4±0.9	0.717
LV mass index, g/m2	82±12	66±9	<0.0001
LVEDVi, mL/m2	99±13	83±12	<0.0001
LVESVi, mL/m2	38±8	33±10	<0.01
LVSVi, mL/m2	61±8	50±8	<0.0001
RVEF, %	57±7	59±7	0.174
RVEDVi, mL/m2	102±16	85±14	<0.0001
RVESVi, mL/m2	44±11	35±10	<0.0001
RVSVi, mL/m2	58±11	50±8	<0.001
LGE present	15 (19)	0 (0)	<0.05

Numbers are mean±SD for continuous and n (%) for categorical data. BMI, body mass index; BSA, body surface area; BP, blood pressure; HR, heart rate; hs-TNT, high-sensitivity troponin T; LGE, late gadolinium enhancement; LV, left ventricular; LVEF, left ventricular ejection fraction; LVEDVi, left ventricular end-diastolic volume index; LVESVi, left ventricular end-systolic volume index; LVSVi, left ventricular stroke volume index; NT-proBNP, N-terminal pro-B-type natriuretic peptide; RVEF, right ventricular ejection fraction; RVEDVi, right ventricular end-diastolic volume index; RVESVi, right ventricular end-systolic volume index; RVSVi, right ventricular stroke volume index; V̇O_2peak, peak oxygen uptake.

Presence, Localization and Extent of Myocardial Fibrosis on Contrast-Enhanced CMR

CMR revealed non-ischemic LGE in 15 of the 78 (19%) male triathletes, but in none of the male controls (P<0.05). A LGE pattern typical of myocarditis was found in 13 triathletes, including 7 triathletes with subepicardial LGE location and a thin pericardial gap, while 6 triathletes had mid-myocardial LGE. Two athletes had LGE located at the posterior RV insertion point. LGE was typically located in the inferolateral segments (Figures 1B,3D). The anterior wall and the apex revealed no fibrosis. Global LGE size was 2.9±2.3 %LV (range 0.5–9.2%) and global LGE mass of 2.2±2.0 g/m² (range 0.3–7.4 g/m²). Segmental LGE size was 13.6±13.1% per segment with LGE (range 0.3–53% per segment with LGE) in 38 LGE+ segments.

Demographic and CMR Characteristics of LGE+ and LGE− Triathletes

LGE+ triathletes showed a trend for a higher age (P=0.096) and weight (P=0.071) and had higher BMI (P<0.05) compared with LGE− triathletes (Table 2). Peak systolic blood pressure was higher in LGE+ triathletes, at 221±21 mmHg, compared with LGE− triathletes (193±27 mmHg, P<0.001). V̇O_2peak and maximal power were similar for both groups, as were blood biomarkers of myocardial damage, including hs-TnT and NT-proBNP (Table 2). The training history revealed no differences between the groups of triathletes. EF was identical in both groups of triathletes. LGE+ triathletes had higher LV mass index at 89±12 g/m² compared with LGE− triathletes at 81±12 g/m² (P<0.05).

Table 2. Demographic and CMR Parameters in LGE+ and LGE− Male Triathletes

	LGE+ male triathletes (n=15)	LGE− male triathletes (n=63)	P value
Demographic parameters
Age, years	47±9	42±11	0.096
Weight, kg	82±8	77±8	0.071
Height, m	1.82±0.05	1.82±0.07	0.926
BMI, kg/m2	24.7±2.5	23.4±2.2	<0.05
BSA, m2	2.03±0.11	1.98±0.13	0.188
Exercise parameters
Systolic BP at rest, mmHg	131±17	123±14	0.106
Diastolic BP at rest, mmHg	83±11	83±10	0.893
Peak systolic BP, mmHg	221±21	193±27	<0.001
Peak diastolic BP, mmHg	89±12	91±21	0.728
HR at rest, beats/min	66±13	67±13	0.658
Peak HR, beats/min	165±11	170±13	0.242
ΔHR rest/peak, beats/min	100±14	104±16
V̇O2peak, mL/kg per min	53±8	54±10	0.741
Maximal power, W	375±79	421±110	0.146
Blood parameters
Hematocrit, %	0.44±0.46	0.44±0.25	0.217
hs-TNT, pg/mL	7±5	7±9	0.734
NT-proBNP, pg/mL	63±125	35±29	0.107
Training history
Training hours per week	11±3	11±4	0.933
Active years, n	15±7	13±8	0.329
CMR parameters
HR at CMR, beats/min	53±9	53±8	0.931
LVEF, %	62±6	62±5	0.958
Cardiac index, L/min/m2	3.3±0.6	3.3±0.7	0.860
LV mass index, g/m2	89±12	81±12	<0.05
LVEDVi, mL/m2	96±13	100±13	0.332
LVESVi, mL/m2	36±7	38±9	0.492
LVSVi, mL/m2	59±11	62±7	0.400
RVEF, %	59±7	57±6	0.247
RVEDVi, mL/m2	101±19	103±16	0.670
RVESVi, mL/m2	41±10	45±11	0.230
RVSVi, mL/m2	60±14	58±9	0.512
Global LGE size, %LV	2.9±2.3
Global LGE mass, g/m2	2.2±2.0
Segmental LGE size, % per segment	13.6±13.1

Numbers are mean±SD for continuous and n (%) for categorical data. Abbreviations as in Table 1.

Reproducibility of Myocardial Strain Measurement

Bland-Altman analysis revealed very good reproducibility of the longitudinal, radial and circumferential global and segmental strain measurements with high interobserver agreement and without systematic over- or underestimation, narrow absolute and relative limits of agreement and high intraclass correlation coefficients of 0.91–0.99 (Figure 2, Supplementary Table).

Figure 2.

Schematic Bland-Altman graphs of radial, longitudinal and circumferential segmental strain revealing good agreement between the 2 observers without systematic over- or underestimation, low relative differences and narrow limits of agreement.

Segmental Strain in Controls

Segmental strain analysis showed regional differences in longitudinal and radial strain in the male controls (Figure 3). The most contraction with highest radial and longitudinal strain was observed in the controls in the basal inferolateral segments (e.g., segments 4 and 5), which were typically affected by fibrosis in LGE+ triathletes. For example, 80% of LGE+ triathletes had fibrosis in segment 5 and 60% of LGE+ triathletes had fibrosis in segment 4 (Figure 3D).

Figure 3.

Segmental radial (A), longitudinal (B) and circumferential (C) strain analyses in controls revealed regional differences, with the highest radial and longitudinal strain in the inferolateral segments. (D) Frequency of segmental fibrosis in LGE+ triathletes. For example, 80% of LGE+ triathletes had fibrosis in segment 5 and 60% of LGE+ triathletes had fibrosis in segment 4. These 2 segments had the most radial and longitudinal strain in normal subjects. Segments with fibrosis are marked in red. LGE, late gadolinium enhancement.

Global and Segmental Myocardial Strain in LGE+ and LGE− Triathletes

Global radial strain was lower in LGE+ triathletes, with 40±7%, compared to LGE− triathletes (45±7%, P<0.05) and controls (45±8%, P<0.05, Table 3). Global longitudinal strain did not differ among the LGE+ and LGE− triathletes and controls. Global circumferential strain was lower in the LGE+ and LGE− triathletes compared with controls (P<0.05 for both groups), but there were no differences between the LGE+ and LGE− triathletes. Segmental strain analysis showed that radial strain was significantly lower in 4 segments of which 2 segments were typically affected by LGE (i.e., segments 5 and 6) and 2 were directly adjacent to the LGE+ segments (i.e., segments 7 and 13) in LGE+ triathletes compared with LGE− triathletes (Table 3, Figure 4). For longitudinal strain, segments 3 and 4 had lower strain in LGE+ triathletes compared with LGE− triathletes and for circumferential strain, segments 15 and 16 had lower strain in LGE− triathletes compared with LGE+ triathletes (Table 3, Figure 4).

Table 3. Global and Segmental Strain in Male Controls, LGE+ and LGE− Triathletes

Strain, %	Radial strain				Longitudinal strain				Circumferential strain
	Controls	LGE+	LGE−	P value	Controls	LGE+	LGE−	P value	Controls	LGE+	LGE−	P value
Global strain	45±8	40±7†	45±7*	0.053	−18±2	−17±2	−18±2	0.141	−18±4	−15±4†	−14±3#	<0.0001
Basal segmental strain
1 Anterior	33±10	33±13	33±13	0.999	−15±7	−15±5	−16±5	0.645	−21±8	−18±6	−18±5ll	0.153
2 Anteroseptal	26±11	23±9	24±13	0.770	−16±9	−16±5	−16±7	0.965	−20±4	−16±5†	−17±4ll	<0.05
3 Inferoseptal	17±8	18±6	17±9	0.976	−26±7	−21±6‡	−25±6*	<0.05	−19±7	−13±4‡	−14±4¶	<0.01
4 Inferior	55±16	53±16	57±18	0.738	−38±8	−34±8†	−37±7*	0.161	−12±4	−9±4†	−9±4¶	<0.01
5 Inferolateral	65±19	47±17§	56±14*,#	<0.01	−41±10	−42±8	−40±9	0.836	−17±7	−15±5	−13±5¶	<0.05
6 Anterolateral	66±16	54±11‡	62±13*	<0.05	−31±6	−32±8	−32±8	0.636	−17±6	−13±7	−13±5¶	<0.01
Middle segmental strain
7 Anterior	46±14	43±15	52±15*,ll	0.055	−22±7	−21±7	−22±5	0.924	−19±5	−17±3	−17±5ll	0.095
8 Anteroseptal	28±11	29±11	31±12	0.621	−17±5	−15±7	−16±5	0.568	−19±5	−16±5	−18±4	0.105
9 Inferoseptal	35±8	33±7	34±8	0.833	−11±5	−11±6	−10±6	0.644	−20±7	−17±4	−15±4#	<0.001
10 Inferior	46±12	41±9	44±11	0.420	−14±5	−12±4†	−11±5ll	0.050	−14±5	−11±5	−9±3#	<0.0001
11 Inferolateral	53±11	43±12‡	50±11	<0.05	−16±7	−14±6	−14±6	0.380	−16±5	−13±4	−13±5¶	<0.05
12 Anterolateral	49±8	43±10	48±10	0.150	−14±6	−12±6	−13±5	0.481	−14±4	−9±3†	−9±4¶	<0.01
Apical segmental strain
13 Anterior	52±13	42±13†	50±12*	<0.05	−15±4	−16±4	−15±4	0.663	−21±6	−17±4†	−16±5#	<0.01
14 Anteroseptal	39±13	32±11	38±13	0.151	−17±5	−17±4	−18±5	0.382	−25±7	−21±8	−19±5#	<0.01
15 Inferoseptal	52±14	49±11	54±11	0.293	−12±4	−11±3	−12±5	0.490	−21±8	−17±9	−14±6*,#	<0.001
16 Inferior	51±12	43±9†	50±11	0.059	−12±4	−11±4	−12±4	0.572	−18±6	−15±6	−12±6*,#	<0.01

*P<0.05 for LGE+ vs. LGE−; ^†P<0.05 or ^‡P<0.01 or ^§P<0.001 for LGE+ vs. Controls; ^llP<0.05 or ^¶P<0.01 or ^#P<0.001 for LGE− vs. Control. LGE, late gadolinium enhancement.

Figure 4.

Segmental strain analysis in LGE+ and LGE− triathletes. Radial strain was significantly lower in segments 5, 6, 7 and 13 in LGE+ triathletes compared with LGE− triathletes. Longitudinal strain was lower in segments 3 and 4, while circumferential strain was lower in segments 15 and 16 in LGE+ triathletes compared with LGE− triathletes. LGE, late gadolinium enhancement.

Correlation Between Segmental LGE Size and Segmental Strain

An inverse correlation was found between segmental LGE size and relative radial strain (P<0.01, Figure 5A), indicating the extent of myocardial fibrosis has a direct effect on myocardial contractility. No correlation was found between segmental LGE size and the relative longitudinal and circumferential strains (P=0.302 and P=0.840, respectively; Figure 5B,C).

Figure 5.

Correlation between segmental LGE size and segmental radial (A), longitudinal (B) and circumferential (C) strain. An inverse correlation was found between LGE size and relative radial strain (A, P<0.01), indicating that the extent of myocardial fibrosis has a direct effect on myocardial contractility. No correlation was found between LGE size and relative longitudinal and circumferential strain. LGE, late gadolinium enhancement.

Discussion

The current study analyzed systolic LV function using feature-tracking CMR strain analysis in competitive LGE+ and LGE− male triathletes and matching controls, providing insights into the potential mechanisms of myocardial fibrosis in athletes. First, we found regional differences in segmental myocardial strain in controls, with the highest strain in the inferolateral segments, which were typically affected by myocardial fibrosis in LGE+ triathletes. This finding suggests a hypothesis that increased contraction of the inferolateral myocardium could be an important co-factor in the development of myocardial fibrosis in competitive triathletes. Second, global and segmental radial strain was reduced in LGE+ compared with LGE− triathletes, but standard cine CMR revealed normal LV function in all triathletes, demonstrating the higher sensitivity of strain analysis for detecting abnormal systolic LV function. Third, an inverse correlation was found between LGE size and relative radial strain, revealing the direct effect of LGE size on myocardial function. Fourth, global circumferential strain was similarly reduced in LGE+ and LGE− triathletes compared with controls, indicating that this reduction represents a normal physiological adaption of the athlete’s heart rather than a pathological finding.

Regional Differences in Segmental Myocardial Strain in Controls

We found a distinct pattern of segmental longitudinal and radial strain in the controls, with the highest contraction in the basal inferolateral segments. A similar pattern was found by Tang et al, who analyzed segmental strain in healthy controls and in patients with ventricular arrhythmia without structural heart disease using feature-tracking CMR.²² The increased myocardial contraction of the inferolateral myocardial segments in our controls could be one explanation for the observed predominant occurrence of non-ischemic fibrosis in this myocardial region. Our data may generate the hypothesis that increased contraction of the inferolateral myocardium could be an important co-factor in the development of myocardial fibrosis in competitive triathletes, who are exposed to a repetitive and sustained increase in myocardial wall stress during training and competition. LGE+ triathletes characteristically had higher systolic blood pressure under exercise and higher LV mass, resulting in an increased myocardial oxygen demand. The coincidence of physiologically increased contraction of the inferolateral wall, adaptive LV hypertrophy and pathologically increased blood pressure seem to be a likely explanation for the observed frequent LGE involvement of the inferolateral wall in triathletes.

Reduced Strain in LGE+ Triathletes With Normal EF

We observed focal myocardial LGE in 19% of male competitive triathletes, which is in line with previous studies demonstrating an incidence of myocardial fibrosis in 0–50% of asymptomatic athletes.^1–7 Our data showed reduced global radial and segmental strain in LGE+ triathletes with normal LV function on standard cine CMR, revealing the higher sensitivity of strain analysis for detecting abnormal systolic LV function. A similar finding was observed by Kraigher-Krainer et al, who used strain analysis to study patients with symptoms of heart failure and preserved EF >55%.¹² That study showed a reduced global strain in those patients compared with both patients with hypertensive heart disease and controls. They concluded that abnormalities of LV systolic function measured by strain analysis may contribute to the syndrome of heart failure with preserved EF. Our study further highlighted the potential of strain analysis. EF was unable to reveal the consequences of subtle myocardial fibrosis in LGE+ triathletes, whereas radial strain analysis revealed reduced systolic LV contraction in these subjects. Therefore, strain analysis appears to be more suited to studying the effect of subtle myocardial fibrosis on systolic function.

Correlation Between Segmental LGE Size and Myocardial Strain

We observed an inverse correlation between segmental LGE size and segmental radial strain reduction, revealing the effect of LGE size on regional systolic LV function because radial strain represents myocardial wall thickening.²³ No correlation was found between longitudinal or circumferential strain and LGE size, which is most likely related to the fact that these strains represent different components of myocardial contraction. Longitudinal strain reflects LV shortening from apex to basis, whereas circumferential strain represents myocardial fiber shortening along the circular perimeter.²³ Our data showed that both components of myocardial contraction are less influenced by subtle LGE, because we did not observe differences in global longitudinal and circumferential strain between LGE+ and LGE− triathletes. A similar finding was reported by Maret et al,²⁴ who compared radial and longitudinal strain by feature-tracking CMR with infarct size transmurality by LGE-CMR. That study showed an almost linear decline of radial strain in segments with increasing infarct size transmurality, but no such correlation was found for longitudinal strain.

In the present study, global circumferential strain was similarly reduced in the LGE+ and LGE− triathletes compared with controls, indicating that this reduction represents a normal physiological adaption of the athlete’s heart rather than a pathological finding. This assumption is supported by the fact that our triathletes had larger LVSVi at lower heart rates compared with the controls, resulting in identical cardiac output. This adaption reveals the capacity of the triathlete’s heart to produce normal cardiac output at lower circumferential strain. Our finding is supported by a recent study of Swoboda et al, who revealed reduced circumferential but similar longitudinal strain by tagging CMR in endurance athletes compared with controls.²⁵

Study Limitations

Because of the small sample size we could not analyze the relative contribution of the factors that potentially promote the occurrence of myocardial fibrosis in triathletes, such as increased blood pressure under exercise, increased myocardial mass and regional differences in myocardial contraction. Currently, no echocardiographic strain data are available to further support the current findings on reduced global or regional strain in LGE+ triathletes.

Conclusions

The observed normal regional differences in segmental myocardial strain with the highest contraction in the inferolateral segments may explain the typical myocardial fibrosis of this region in triathletes. Furthermore, radial strain analysis revealed reduced systolic function in LGE+ triathletes, revealing a negative effect of myocardial fibrosis on LV contraction in asymptomatic triathletes with normal EF. Follow-up studies are warranted to analyze the long-term effects of reduced strain on myocardial function measured by EF and on development of clinical symptoms of heart failure.

Disclosures

Dr. Stehning is an employee of Philips Healthcare, Hamburg, Germany. The other authors report no conflicts.

Supplementary Files

Please find supplementary file(s);

http://dx.doi.org/10.1253/circj.CJ-18-1388

References

• 1.   Breuckmann F, Möhlenkamp S, Nassenstein K, Lehmann N, Ladd S, Schmermund A, et al. Myocardial late gadolinium enhancement: Prevalence, pattern, and prognostic relevance in marathon runners. Radiology 2009; 251: 50–57.
• 2.   Wilson M, O’Hanlon R, Prasad S, Deighan A, Macmillan P, Oxborough D, et al. Diverse patterns of myocardial fibrosis in lifelong, veteran endurance athletes. J Appl Physiol 2011; 110: 1622–1626.
• 3.   La Gerche A, Burns AT, Mooney DJ, Inder WJ, Taylor AJ, Bogaert J, et al. Exercise-induced right ventricular dysfunction and structural remodelling in endurance athletes. Eur Heart J 2012; 33: 998–1006.
• 4.   Bohm P, Schneider G, Linneweber L, Rentzsch A, Krämer N, Abdul-Khaliq H, et al. Right and left ventricular function and mass in male elite master athletes: A controlled contrast-enhanced cardiovascular magnetic resonance study. Circulation 2016; 133: 1927–1935.
• 5.   McDiarmid AK, Swoboda PP, Erhayiem B, Lancaster RE, Lyall GK, Broadbent DA, et al. Athletic cardiac adaptation in males is a consequence of elevated myocyte mass. Circ Cardiovasc Imaging 2016; 9: e003579.
• 6.   Abdullah SM, Barkley KW, Bhella PS, Hastings JL, Matulevicius S, Fujimoto N, et al. Lifelong physical activity regardless of dose is not associated with myocardial fibrosis. Circ Cardiovasc Imaging 2016; 9: e005511.
• 7.   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: 111–117.
• 8.   van de Schoor FR, Aengevaeren VL, Hopman MTE, Oxborough DL, George KP, Thompson PD, et al. Myocardial fibrosis in athletes. Mayo Clin Proc 2016; 91: 1617–1631.
• 9.   Tahir E, Starekova J, Muellerleile K, von Stritzky A, Münch J, Avanesov M, et al. Myocardial fibrosis in competitive triathletes detected by contrast-enhanced CMR correlates with exercise-induced hypertension and competition history. JACC: Cardiovasc Imaging 2018; 11: 1260–1270.
• 10.   Moon JC, Sheppard M, Reed E, Lee P, Elliott PM, Pennell DJ. The histological basis of late gadolinium enhancement cardiovascular magnetic resonance in a patient with Anderson-Fabry disease. J Cardiovasc Magn Reson 2006; 8: 479–482.
• 11.   Claus P, Omar AMS, Pedrizzetti G, Sengupta PP, Nagel E. Tissue tracking technology for assessing cardiac mechanics: Principles, normal values, and clinical applications. JACC: Cardiovasc Imaging 2015; 8: 1444–1460.
• 12.   Kraigher-Krainer E, Shah AM, Gupta DK, Santos A, Claggett B, Pieske B, et al. Impaired systolic function by strain imaging in heart failure with preserved ejection fraction. J Am Coll Cardiol 2014; 63: 447–456.
• 13.   Arenja N, Riffel JH, Fritz T, André F, aus dem Siepen F, Mueller-Hennessen M, et al. Diagnostic and prognostic value of long-axis strain and myocardial contraction fraction using standard cardiovascular MR imaging in patients with nonischemic dilated cardiomyopathies. Radiology 2017; 283: 681–691.
• 14.   Romano S, Judd RM, Kim RJ, Kim HW, Klem I, Heitner JF, et al. Feature-tracking global longitudinal strain predicts death in a multicenter population of patients with ischemic and nonischemic dilated cardiomyopathy incremental to ejection fraction and late gadolinium enhancement. JACC: Cardiovasc Imaging 2018; 11: 1419–1429.
• 15.   Schulz-Menger J, Bluemke DA, Bremerich J, Flamm SD, Fogel MA, Friedrich MG, et al. Standardized image interpretation and post processing in cardiovascular magnetic resonance: Society for Cardiovascular Magnetic Resonance (SCMR) board of trustees task force on standardized post processing. J Cardiovasc Magnetic Reson 2013; 15: 35.
• 16.   Cerqueira MD, Weissman NJ, Dilsizian V, Jacobs AK, Kaul S, Laskey WK, et al. Standardized myocardial segmentation and nomenclature for tomographic imaging of the heart: A statement for healthcare professionals from the Cardiac Imaging Committee of the Council on Clinical Cardiology of the American Heart Association. Circulation 2002; 105: 539–542.
• 17.   Barreiro-Pérez M, Curione D, Symons R, Claus P, Voigt JU, Bogaert J. Left ventricular global myocardial strain assessment comparing the reproducibility of four commercially available CMR-feature tracking algorithms. Eur Radiol 2018; 28: 5137–5147.
• 18.   Morais P, Marchi A, Bogaert JA, Dresselaers T, Heyde B, D’hooge J, et al. Cardiovascular magnetic resonance myocardial feature tracking using a non-rigid, elastic image registration algorithm: Assessment of variability in a real-life clinical setting. J Cardiovasc Magnetic Reson 2017; 19: 24.
• 19.   Heyde B, Bouchez S, Thieren S, Vandenheuvel M, Jasaityte R, Barbosa D, et al. Elastic image registration to quantify 3-D regional myocardial deformation from volumetric ultrasound: Experimental validation in an animal model. Ultrasound Med Biol 2013; 39: 1688–1697.
• 20.   Guazzi M, Arena R, Halle M, Piepoli MF, Myers J, Lavie CJ. 2016 Focused Update: Clinical recommendations for cardiopulmonary exercise testing data assessment in specific patient populations. Circulation 2016; 133: e694–e711.
• 21.   Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1: 307–310.
• 22.   Tang X, Yu S, Yu Y, Ren H, Li S, Zhou L, et al. Left ventricular myocardial strain in ventricular arrhythmia without structural heart disease using cardiac magnetic resonance. Am J Transl Res 2017; 9: 3006–3016.
• 23.   Scatteia A, Baritussio A, Bucciarelli-Ducci C. Strain imaging using cardiac magnetic resonance. Heart Fail Rev 2017; 22: 465–476.
• 24.   Maret E, Todt T, Brudin L, Nylander E, Swahn E, Ohlsson JL, et al. Functional measurements based on feature tracking of cine magnetic resonance images identify left ventricular segments with myocardial scar. Cardiovasc Ultrasound 2009; 7: 53.
• 25.   Swoboda PP, Erhayiem B, McDiarmid AK, Lancaster RE, Lyall GK, Dobson LE, et al. Relationship between cardiac deformation parameters measured by cardiovascular magnetic resonance and aerobic fitness in endurance athletes. J Cardiovasc Magnetic Reson 2016; 18: 48.