Chronic fatigue syndrome in women assessed with combined cardiac magnetic resonance imaging

We studied 48 females with a combined CMR approach. A consecutive series of 12 patients with CFS were recruited from a specialised CFS centre, to be examined with CMR. The diagnosis of CFS was based on the revised case definition by Fukuda et al. [3] after ruling out other potential causal diseases. Measurements in the 12 CFS patients were then compared with 36 age-matched, healthy female volunteers without any comorbidities (control group), who provided informed consent prior to CMR examination. The study complied with the Declaration of Helsinki for investigation in human beings, and was performed after approval and supervision of our institutional ethics committee.

CMR examination was performed on a 1.5-T whole body scanner (Achieva scan, Philips Medical Systems, Best, the Netherlands) using commercially available cardiac CMR software. For signal reception, a five-element cardiac synergy coil was used. Electrocardiogram triggering was done with a vector electrocardiogram setup. Subjects were examined in the supine position. Morphological images in the standard views were acquired by using fast field echo cine images (slice thickness 8.0 mm; repetition time 3.4 ms; echo time 1.7 ms; flip angle 60°; matrix 256 × 256).

Subsequently, a breath-hold, black-blood, T2-weighted double-inversion recovery sequence with a fat-saturation pulse was performed in 8–10 short-axis slices, with the following parameters: repetition time 1800–2400 ms; echo time 80 ms; matrix 256 × 256; field of view 32–40 cm; slice thickness 12 mm; number of excitations 1. Myocardial scar was assessed on contrast-enhanced multislice standard views, obtained approximately 10 minutes after intravenous bolus injection of 0.2 mmol gadolinium per kilogram body weight (Shering AG, Berlin, Germany). A three-dimensional Turbo Field Echo inversion recovery T1-weighted sequence was used with the following parameters: repetition time 4.0 ms; echo time 1.3 ms; flip angle 15°; inversion time individually optimised to null myocardial signal (usually between 180–250 ms); matrix 157; slice thickness 10–12 mm).

CMR data were analysed on a workstation using dedicated software for cardiac analysis (Philips MR workspace, Release 2.5.3.0 2007-12-03; Philips, the Netherlands). Left ventricular end-diastolic (ml) and end-systolic volumes (ml), stroke volume (ml), LVEF (%) cardiac output (l/min), and end-diastolic wall mass (g) were calculated from contiguous short-axis loops by segmentation of endocardial and epicardial borders on each frame. Papillary muscles were regarded as part of the ventricular cavity.

The left ventricular wall regions were divided into 17 segments according to a standardised myocardial segmentation model [13]. Normal wall motion was assigned a score of 0, mild hypokinesia 1, severe hypokinesia 2, akinesia 3, and dyskinesia as 4. The wall motion score index was calculated by dividing the sum of scores in each segment by the total number of observed segments. T2-weighted images were considered if abnormal increased signal intensity was visually observed within the myocardium.

The contrast-enhanced images were independently evaluated by three experienced observers, who were blinded to patient/control characteristics and to the evaluations of each other. Contrast enhancement was defined as a zone of hyperenhancement on the late contrast-enhanced images (in contrast with the dark-grey signal of the normal myocardium). The studies were considered abnormal when at least two independent observers described the same abnormalities with agreement in both the presence of contrast enhancement and its location (same segment), reproducible in more than one contiguous slice and in more than one projection.

Twelve female CFS patients and 36 age- and gender-matched controls were analysed in this study. The patients and controls were relatively young (36 ± 13 vs. 29 ± 8 years). All but one CFS patient had a history of infectious mononucleosis disease with IgG antibodies to Epstein-Barr virus or human cytomegalovirus capsid antigen; all of these eleven patients had antibodies to Epstein-Barr virus nuclear antigen. Demographics and baseline characteristics did not differ between both groups (see also Table 1).

The complete CMR examination protocol was followed in all CFS patients and controls; image sequences of a high quality could be obtained in all cases. CMR data of our CFS population are presented in Table 2. In CFS patients, LVEF (57.9 ± 4.3 %; p < 0.01), end-diastolic diameter (44 ± 3.7), as well as body surface area corrected end-diastolic LV volume (77.5 ± 6.2 ml/m²; p < 0.01), LV stroke volume (44.9 ± 4.5 ml/m²; p < 0.001), and LV wall mass (39.8 ± 6.5 g/m²; p = 0.02) were significantly lower than age- and gender-matched controls. Mild wall motion abnormalities were observed in four CFS patients in the basal and/or mid inferoseptal wall, leading to a wall motion score index of 0.02 ± 0.04; none of the controls showed wall motion abnormalities. Myocardial damage (i. e. fibrosis), as indicated by the presence of contrast enhancement, was observed in three patients who showed such lesions in the basal inferoseptal, septal, and anteroseptal midwall segments. An example is presented in Fig. 1. No contrast enhancement was seen in healthy controls. Only one patient showed both contrast enhancement and wall motion abnormalities. Regional or global oedema, as indicated by an increased signal intensity on T2-weighted images, was not observed in any of the patients.

The present CMR study used a combined CMR approach of cine, contrast enhancement, and T2-weighted imaging to assess cardiac dimensions, function, and myocardial tissue characteristics in patients with confirmed CFS and healthy controls. Our data demonstrate a left ventricular size, mass, and function that were lower than in age- and gender-matched controls, although still within the normal range. Theoretically, this difference could partially explain some mild fatigue in these patients, but it may also be suggested that the reduction in LV dimensions and function is the result of reduced physical activity. In this respect, Takenaka et al. found significantly reduced ventricular diastolic dimensions and cardiac output after 20 days of bed rest [14]. These findings are also in accordance with a previous echocardiographic study, which revealed a smaller LV chamber with a lower cardiac output in CFS patients during pharmacological stress testing [15].

Recently, Hollingsworth et al. investigated geometrical and functional LV parameters with CMR, including strain analysis, in 12 CFS patients and 10 healthy controls [7]. Although not all analyses gave evidence of myocardial dysfunction in CFS patients, residual torsion at 150 % of the end-systolic time was greater in CFS patients than controls, suggesting a delay in the release of torsion. In CFS patients, both residual torsion at 150 % of the end-systolic time and torsion to endocardial strain ratio correlated negatively with the end-diastolic volume index. These findings may contribute to the understanding of cardiac functional impairments in CFS patients [7].

In the present study, we also performed contrast-enhanced imaging in both CFS patients and a control group. We found no increase in signal intensity (i. e. oedema) on the T2-weighted CMR images, which demonstrates that our patients did not suffer from an active state of viral myocardial infection at the time of CMR. Midwall contrast enhancement, a finding that has frequently been observed in patients with histopathological evidence of chronic active or borderline myocarditis [16], was found in three out of 12 CFS patients. As demonstrated by Mahrholdt et al. in patients with acute myocarditis, contrast enhancement may not be detectable in approximately 25 % of patients at an average follow-up of 4.5 months [17]. Therefore, one may hypothesise that the initial myocardial involvement in our CFS patient population might have been somewhat larger than observed in our CMR examination during chronic state. The presence of a non-permissive, persistent viral infection, in which only a very low level of complete infectious virus is produced, could be a potential explanation of isolated midwall contrast enhancement (so, in the absence of oedema) [18] as observed in three of our CFS patients.

Previous myocarditis CMR studies demonstrated a relation between the type of virus and the pattern of myocardial damage [17]. Accordingly, the distribution of contrast enhancement may help to distinguish between different viral infections of the myocardium. In our CFS patients, contrast enhancement was exclusively observed in the infero- and antero-septum, which fits well with myocardial infections by Epstein-Barr virus [19], or human cytomegalovirus; serum antibodies for at least one of these viruses were identified in 11 of our CFS patients. Notably, inferoseptal wall motion abnormalities were observed in one patient with and three patients without contrast enhancement. In patients with myocarditis, the value and interpretation of contrast enhancement can be difficult because of the heterogeneous scar distribution and the generally lower signal intensities. In addition, due to limited voxel resolution, smaller myocardial scars may not be detected. Another explanation for this discrepancy could be the absence of contrast enhancement in the chronic state. To the best of our knowledge, this study is the first to suggest a septal predilection pattern of contrast enhancement and/or wall motion abnormalities in CFS patients. In CFS patients, myocardial fibrosis may lead to ventricular dysfunction, as in patients with Chagas myocarditis, for instance, in whom a relation between the amount of contrast enhancement and late ventricular dysfunction has been described [20].

In the present study, in patients with CFS, diagnosis was based on the diagnostic criteria of Fukuda et al. [3]. Future studies might also assess cardiac morphology and function in patients with myalgic encephalomyelitis and its diagnostic criteria [21]. Few data are available on prevalence and extent of cardiac involvement in CFS patients. With our combined CMR approach, we demonstrated in one examination (although in the normal range) lower size, mass, and function of the left ventricle and structural abnormalities of the myocardium, as previously assessed separately with different techniques. Specific treatment of myocarditis remains a challenge [6, 22]. A small, randomised, placebo-controlled trial on the use of antiviral therapy for patients with a Epstein-Barr virus subset of CFS found a clinical improvement after six months [23]. Recently, it has been suggested that Epstein-Barr virus can provoke ventricular tachycardia by both acute and chronic myocardial inflammation, emphasising the clinical importance of this viral infection [19]. To test the hypothesis that Epstein-Barr and/or human cytomegalovirus may lead to ventricular dysfunction, myocardial involvement, and CFS, further studies are warranted that should include early and repeat examinations with a combined CMR approach plus targeted myocardial tissue biopsies. Such serial studies may answer the question of whether some CFS patients develop a dilated cardiomyopathy with the inherent risk of an inferior clinical course [6].

We did not perform serial serological tests or myocardial biopsies. It would have been ideal to correlate CMR findings with targeted endomyocardial biopsies. However, this would have involved an invasive procedure without a direct therapeutic implication. In patients with myocarditis, the value and interpretation of contrast enhancement can be difficult because of heterogeneous scar distribution and the generally lower signal intensities [4, 24‐26]. Although our data are unique, findings in this small population of CFS patients cannot answer the question whether the somewhat lower cardiac function may contribute to the reduced exercise capacity or may be a result of lower physical activity. Further assessment of relationships between exercise capacity and CMR parameters in a larger patient population is of interest and should be addressed by future studies.