Comparative analyses of muscle MRI and muscular function in anti-synthetase syndrome patients and matched controls: a cross-sectional study

Data on muscle MRI findings and muscular outcome in ASS are limited. Here we showed that MRI findings compatible with either muscle damage (i.e. fatty replacement and/or reduction in muscle volume) or active myositis (edema in muscle and/or fascia) were present in 65% of the patients in an ASS cohort with a median 6 years of disease duration. Highly significant differences between patients and healthy controls were seen in all MRI muscle parameters, muscle strength and endurance.

Muscle MRI has been performed in a number of different muscle conditions, using semiquantitative and quantitative methods [30–34]. Taken together, these data indicate that some considerations should be made when evaluating muscle MRI in myositis. First, muscle MRI changes do not have to represent myositis. In this study, we also identified edema, fatty replacement and volume reduction in the healthy controls. However, the MRI changes in healthy controls were much less pronounced than in the patients; only three controls had a muscle edema score??4 and only one control had concomitant signs of muscle edema and damage. Two controls had edema evaluated as post-traumatic changes, a well-known cause of MRI muscle edema [35]. The four controls with total MRI score???10 were all over 70 years of age. This is relevant because the total MRI score in the controls correlated with age at study inclusion. Second, muscle MRI changes can also be influenced by gender, especially muscle volume. Because no definitions of normal muscle volumes exist, muscle atrophy is difficult to assess. By using healthy controls matched for gender, and a method which would underestimate rather than overestimate muscle atrophy, we tried to reduce this problem of interpretation. However, we still believe that the data on muscle volume reduction must be interpreted with some caution. Interestingly, we discovered a high frequency of fatty streaks (Gouttalier score?=?1) in muscle compartments both in patients (33%) and in controls (40%). Not surprisingly, the frequency of fatty streaks (and the fatty replacement score) also correlated with age at study inclusion in both groups, possibly indicating that a “minor extent” of fatty replacement could be a natural finding in muscle anatomy.

The current study did not reveal any specific ASS-related MRI pattern equivalent to the recently described ASS-related muscle histology pattern with necrotic myofibers in the perimysium and inflammation in adjacent connective tissue [21, 22]. However, some potentially interesting observations were made. First, we found that 16 patients (eight PM and eight DM) were described with fascial edema and 13/16 of these had concurrent muscle edema. Previously, fascial edema has mostly been described in DM and found to correlate with histopathological findings [13, 36]. We speculate that the observed MRI pattern with concurrent edema in muscle and fascia is a feature of active ASS, regardless of PM/DM subset, but this needs to be evaluated in further studies. Second, we found a trend toward partition of the MRI findings, with predominance of fatty replacement in the posterior compartment and muscle edema in the anterior compartment. Because this study had a cross-sectional design, we do not know whether the more pronounced damage in the posterior musculature was a consequence of relatively early and/or intense inflammation in this compartment. Interestingly, fatty replacement (Gouttalier score???2) was also present in a small number of the controls (N?=?3), all in the posterior part of the thighs, so these findings need to be further investigated. A similar MRI pattern, with edema anteriorly and fatty replacement posteriorly, has recently been described in PM patients, including patients with immune-mediated necrotizing myopathy (IMNM) [14, 30].

The total edema score correlated with CK levels. This is consistent with recent data from Italy where muscle edema and CK correlated in 51 IIM patients [37]. However, in contrast to other studies [37, 38], there was no correlation between the three muscle tests evaluated and the total edema score. Instead, we found that muscle strength correlated with the total damage score. Hence, it is possible that the current study design, with cross-sectional evaluation of patients with established disease, is most suitable for MRI assessment of chronic muscle changes. It should, however, be noted that 26% of the ASS patients with visible muscle damage had concurrent signs of activity (edema), and that 23% of the ASS patients, all with normal CK values, had a total edema score???43% of maximum. Because muscle edema was also seen in 12% of the controls (all but one with normal CK values), one could discuss the clinical importance of this finding. It should, however, be noted that the extent and intensity of edema in the controls was far less than in the patients. In fact, the 90th centile of total edema score in the controls was only 4, compared to 26 in the patients. In addition, the fact that the ASS patients with no previous myositis diagnosis (N?=?13) had higher MRI scores than their correspondent healthy control (p??0.02) emphasizes the importance of muscle MRI to evaluate the possibility of late-onset/subclinical myositis in ASS.

It has earlier been described that anti-Ro52 is associated with poor outcome in ASS-related ILD [39, 40], and Marie et al. [41] have also reported anti-Ro52 as a prognostic factor for the muscle component in the syndrome. This was also seen in the current study, independently of anti-Jo1 status (p??0.05). Hence, anti-Ro52 seems to be a possible prognostic factor for both the lung and muscle components in ASS. Furthermore, subgroup analyses showed an association between anti-Jo1 positivity and total edema score in the patients with earlier diagnosed myositis. These findings should be evaluated carefully, however, since the anti-Jo-1-negative myositis group was small (N?=?5).

The study has some potential limitations. First, the cross-sectional study design limits evaluation of potential predictive muscle MRI variables and the progression of these. Second, muscle endurance was also tested in muscle groups other than thigh muscles; it is therefore possible that whole-body MRI could correlate better with the FI2 test. Third, we did not specifically investigate possible associations between different kinds of treatment, MRI findings and muscular functions in the ASS patients.