Myocardial fibrosis in athletes—Current perspective

Abstract Several previous studies suggested that prolonged and extensive physical activity might lead to increased prevalence of myocardial fibrosis in athletes. The review summarizes these studies focusing on common patterns of myocardial fibrosis observed in athletes, their potential causes and significance. It also presents recent research on parametric imaging shedding new light on diffuse myocardial fibrosis in athletes. Finally, it reviews how these traditional and novel cardiac magnetic resonance (CMR) techniques can be incorporated in the diagnostic work up to differentiate athlete's heart from cardiomyopathies.


| INTRODUCTION
Studies have shown that prolonged endurance exercise leads to marked elevation of myocardial necrosis markers, brain natriuretic peptides and an increase of inflammatory markers. [1][2][3] The raise of biochemical markers is directly proportional to the duration and intensity of exercise and inversely proportional to the amount of prior training. 1 Some imaging heart studies, but not all, 4 have shown that biochemical changes were accompanied by a decrease of cardiac performance, particularly in relation to the right ventricle 5,6 and by signs of myocardial oedema. 6 These alterations were shown to be only transient and do not persist in a longer perspective. 5,6 Importantly, none of those studies have shown any myocardial fibrosis related to the studied endurance events. [4][5][6] It was suggested that increase of troponin levels is due to release of cytosolic fraction of this marker without compromise to cardiomyocates or that it is related to muscular fatigue or even transient renal impairment and decreased troponin clearance. 1 However, other studies demonstrated that some lifelong endurance athletes might be at increased risk of ventricular arrhythmias originating from the right ventricle. 7,8 These observations were followed by a group of studies suggesting increased prevalence of fibrosis in veteran athletes in comparison to sedentary controls of the same age, leading to the hypothesis that prolonged endurance training, especially without adequate recovery may predispose to myocardial fibrosis. 8,9 Apart from the risk of arrhythmias, the consequences of myocardial fibrosis may include increased myocardial stiffness and local cardiac dysfunction. 10 The aim of this review is to summarize current understanding of the possible causes and significance of observed myocardial fibrosis in athletes with new insight from recent studies on diffuse interstitial myocardial fibrosis.

| Focal fibrosis
The detection of myocardial fibrosis with CMR is performed after the administration of a gadolinium-based contrast agent into an antecubital vein. The gadolinium-chelates contrast agents used in CMR are mainly extracellular contrast agents, which have an initial relative short vascular phase, followed by an extracellular phase in which the contrast agent diffuses in the interstitium. There is a typical "washin" and "washout" contrast dynamics in different tissues, including the myocardium. 11 In the normal heart, the interstitial space undergoes normal "washout" of the contrast agent with no contrast accumulation. In the presence of myocardial injury or disease, the extracellular space increases leading to delayed "washout" and contrast accumulation. Higher concentration of contrast agent decreases T1 relaxation time of the studied tissue, thus changing its signal intensity, which appears "bright" (hyperintense) as opposed to the normal myocardium (hypointense). The technique used to image the heart post contrast is called late gadolinium enhancement (LGE) and the images are acquired 10 to 15 minutes postcontrast injection. The two main LGE patterns are (a) "ischemic" type characterized by subendocardial to transmural LGE corresponding to territories of coronary artery supply and (b) "nonischemic" type characterized by midwall or subepicardial location, either focal or patchy, as observed in a range of cardiomyopathies, myocarditis or nonischemic myocardial injury from other causes. While LGE can detect focal myocardial fibrosis, the technique cannot detect interstitial fibrosis.

| Diffuse fibrosis
Recent developments in CMR were focused on introduction to clinical practice of sequences called CMR relaxometry (native T1-mapping and extracellular volume [ECV] of distribution), which is able to detect diffuse fibrosis. 12 This technique is based on mapping of T1-relaxation time of myocardium (measured in milliseconds), which varies in relation to the composition of myocardium and rises with the increase in the amount of fibrotic tissue. Registration of T1-relaxation time maps before and after contrast administration and knowledge of blood viscosity affecting the T1 time (measured by patient's hematocrit) permit calculation of ECV of the myocardium. As native (without contrast) myocardial T1-mapping is believed to vary substantially between various scanners, different contrast agents and T1-mapping sequences, the ECV is believed to adjust for most of these differences and to most reliably depict diffuse fibrosis. [12][13][14] So far, most of the studies on fibrosis in athletes with means of CMR were based only on LGE detection 9,15 but more recently there are studies incorporating T1-mapping and ECV calculation. [16][17][18][19][20][21][22][23] Findings from these studies will be summarized below.

| COMMON PATTERNS OF MYOCARDIAL FIBROSIS IN ATHLETES
Most of the studies on myocardial fibrosis in athletes were based on middle age or veteran endurance athletes, both amateur and professional. 9,15 These athletes are typically between 30 and 60 years of age, predominantly male and have been engaging for at least 5 to 10 years in mainly endurance (running, biking, or combined) exercises.
The presented results should be therefore interpreted in this context. Detailed analysis of studies on LGE detection in athletes has been presented in earlier reviews. 9,15 Therefore, here, only a concise summary of the observations will be provided. In general, in asymptomatic endurance athletes with normal ECG three LGE patterns can be distinguished: two nonischemic and one ischemic. The nonischemic patterns include: (a) mid-myocardial LGE in the insertion points (points in the interventricular septum where right and left ventricle muscles connect called otherwise "hinge points" or junction points) and (b) subepicardial or mid-myocardial LGE in the inferolateral segments or less commonly in the interventricular septum or elsewhere.

| Insertion point fibrosis
Insertion point fibrosis is most often limited to the inferior insertion point ( Figure 1A). It is the most commonly observed pattern in athletes irrespective of age. 9,15 Its prevalence may reach up to 20% to 30% and has been correlated with a cumulative training load and training intensity. 24 These correlations may reflect the time of pressure and/or volume overload present in the right ventricle during intensive exercise, which causes tension on the insertion points and may lead to microinjuries visible later as spots of LGE in that location. 22 In line with this, similar pattern of fibrosis (however more likely to involve both upper and lower insertion points) has been also observed in patients with pulmonary hypertension or pulmonary regurgitation after repair of tetralogy of Fallot. [25][26][27] Importantly, in both of those instances, this type of LGE was shown to be benign.
Another hypothesis suggests that insertion point fibrosis may be related to myocardial hypertrophy visible in some athletes. Accordingly, a similar pattern of fibrosis is also visible in patients with hypertrophic cardiomyopathy, but a large study found that contrary to more pronounced LGE pattern, these small spots of insertion point fibrosis also do not affect prognosis. 28 Insertion point fibrosis has been also observed in around 10% of otherwise healthy elderly individuals and may form one of the elements of an aging heart. 22,29 A recent study has found that in subjects without additional evidence of cardiac damage insertion point fibrosis is to be considered an incidental finding. 30

| Inferolateral or septal nonischemic fibrosis
Inferolateral and septal nonischemic fibrosis is less often observed in athletes than insertion point fibrosis ( Figure 1B). Inferolateral midmyocardial fibrosis has been previously noticed in some rare storage diseases as Fabry disease, 31 but is most characteristic of an acute or healed myocarditis. 11 It is therefore plausible that small, linear subepicardial, or mid-myocardial areas of LGE in the inferolateral segments or in the interventricular septum in asymptomatic endurance athletes reflect previous, usually silent, myocarditis. Intensive, prolonged exercise has been shown to affect immune resistance in the short period after intensive exercise, which if combined with seasonal infections may predispose athletes to myocarditis. 32,33 Although this has been so far shown only in animal models, there are case reports supporting this hypothesis. 34,35 Myocarditis usually resolves without complications and small remnant scars are considered benign and do not require further testing. Such small, silent, mid-wall areas of fibrosis have also been found in almost 4% of general population. 36 Only if they are larger and form striae of LGE, particularly in the anteroseptal location, they may increase the risk of severe arrhythmias and affect prognosis as demonstrated in several studies [37][38][39][40] . However, athletes with large areas of postmyocarditis LGE are more likely to be symptomatic and report palpitations or reduced physical performance or present ECG and echocardiographic changes. [37][38][39] In fact, recent study in a large community-based sample of older adults from Reykjavik has demonstrated that such minor nonischemic fibrosis patterns, as described in asymptomatic athletes, do not influence prognosis when adjusted for traditional risk factors. 41

| Ischemic fibrosis
Ischemic fibrosis has been reported predominantly in athletes above 50 years of age (veteran athletes) and may reflect the lifelong influence of common cardiovascular risk factors present in these individuals ( Figure 1C). Exercise has many beneficial effects on the heart but does not eliminate the presence of all risk factors of atherosclerosis.
In fact, the prevalence of common cardiovascular risk factors in Olympic athletes was surprisingly high including 0.3% of hyperglycemia, 3.8% of hypertension, 8% of smoking habit, 18% of positive family history, 25% of increased waist circumference, and 32% for dyslipidemia. 42 In that study, endurance athletes had generally low cardiovascular risk profile, but one to two risk factors were still present in 50% of them and 2% of them had three to four risk factors. A recent study demonstrated that veteran athletes have more coronary atherosclerotic plaques than age and sex-matched sedentary controls. 43 These plaques may arise from increased shear stress during periods of intensive exercise with elevated coronary pressures or might be caused by transient periods of inflammation. However, the morphology of plaques seems more stable (calcified) with lower prevalence of the soft, vulnerable plaques, which are prone to rupture. 43 Nevertheless, as coronary plaques are present it seems logical that at least some of them may occasionally erode causing small, silent myocardial infarctions ( Figure 1C). In fact, silent myocardial infarctions were found not only in veteran athletes, but also in normal population.
The results of the MESA study have shown subendocardial (ischemic) scars on CMR in 7.9% of people aged 68 ± 9. 36

| NEW DATA FROM PARAMETRIC IMAGING STUDIES
Only a few recent studies on diffuse fibrosis in athletes can be found. [16][17][18][19][20][21][22][23] They are summarized in Table 1. In general, there is a large F I G U R E 1 Most common patterns of late gadolinium enhancement (LGE) observed in athletes. A,Short axis view, mid-myocardial (nonischemic) LGE in the inferior insertion point in an asymptomatic 41-year-old ultramarathon runner without prior medical history (own data), B, three-chamber view, mid-myocardial (nonischemic) LGE in the basal inferolateral segment in an asymptomatic 41-year-old ultramarathon runner without prior medical history (own data), C, short axis view, subendocardial (ischemic) LGE in the mid inferolateral segment in an asymptomatic 52-year-old recreational runner without prior medical history (own data) discrepancy between those studies in terms of scanners and sequences used to analyze T1-mapping, as well as in the use of contrast agent and therefore concomitant LGE reporting (some were based only on native T1-mapping). 17 The native T1 time in athletes varies from 943 to 1268 ms and ECV from 20% to 26%. Therefore, each result should be interpreted in relation to internal control group of healthy individuals and dedicated reference values. [12][13][14] Most of the studies demonstrate that endurance exercise does not lead to an increase of ECV or is related to only slight increase, with absolute values remaining within the reference range for the normal population. [18][19][20][21][22] Two studies reported a decrease of ECV in athletes, which is explained by increase of cardiomyocyte mass rather than ECV in relation to training. 16,23 In one of those studies, ECV inversely correlated with increasing left ventricular mass. 16 Table 2 along with CMR characteristics of the athlete's heart. 44,45 In all conditions LGE detection and its pattern may be of paramount relevance to the diagnosis and management. Recently, also parametric imaging including T1-mapping and ECV calculation have been used to detect diffuse fibrosis in at least some cardiomyopathies such as hypertrophic cardiomyopathy (HCM) and dilated cardiomyopathy. 12

| GAPS IN RESEARCH
Further studies are needed to comprehensively analyze not only the presence, but also the relation between different patterns and amount F I G U R E 2 Most common patterns of late gadolinium enhancement (LGE) observed in athletes with different training history and in sedentary controls in relation to age. All drawings of the left ventricle are in short axis. Three groups are presented: on the left-lifelong athletes who start early in life mainly during adolescence (<20 years of age) and eventually become active or sedentary veteran athletes; in the middle-athletes who start later in life (>20-30 years of age) and continue to become veteran athletes; on the right-sedentary controls. Patterns of fibrosis are described in text and presented in Figure 1. Studies demonstrate that most common patterns of fibrosis in athletes are insertion point fibrosis and myocarditis-type fibrosis. 16,18,19,21,22,24 However, both of those patterns were also found in sedentary individuals. 22,29,30,36 Insertion point fibrosis seems both age and training related and therefore may occur earlier in athletes. 5,22,24 Ischemic fibrosis was occasionally found with similar frequency in veteran athletes 9