Occurrence of focal myositis during Behçet's disease: The identification of a specific vasculitis‐associated focal myopathy

This study aimed to report the association of focal myositis (FM) and Behçet's disease (BD) and to analyse the main characteristics of such an association.


INTRODUCTION
Behçet's disease (BD) is a rare systemic vasculitis characterised by mucocutaneous, ocular, articular, vascular and neurological manifestations, [1] the diagnosis of which is based on clinical signs. [2] Nervous system impairment occurs in 1.3% to 59% of BD cases, mainly involving the central nervous system (CNS). [1] The muscular manifestations of neuro-Behçet's disease are rare and occasionally reported in the literature. [3] Initially described by Heffner et al. in 1977, [4] focal myositis (FM) is defined as a rapidly growing focal inflammatory pseudotumour usually restricted to a single skeletal muscle, preferentially affecting the lower limbs. The prognosis is considered benign with spontaneous regression in most cases. [3] The pathological features include a mixed pattern of myogenic alterations, inflammatory infiltrates and sarcolemmal overexpression of the major histocompatibility complex class (MHC-1). [5] The myogenic features include heterogeneous fibre sizes, rare necrotic fibres or myophagocytosis, and regenerating fibres. Inflammatory infiltrates are mainly composed of macrophages (CD68) and T cells (CD3), which are preferentially CD4 T cells. [5] In a recent study, FM appeared to be frequently associated with other diseases such as immune-mediated inflammatory disease, neoplasia and radiculopathy. [6] Muscular involvement during BD is rarely described, but several case reports of the association of BD and FM have been published. [3,[6][7][8][9][10][11][12][13] The aim of the present study was to describe in more detail the association between BD and FM and to analyse exhaustively the clinical features, pathological data, treatment and outcome of patients, but also to evaluate the specificity of the phenotype (BD + FM+), by comparison with a control group composed of patients with FM but without BD (BD À FM+).

Study design
Case group: FM associated with BD (BD + FM+ group) In this multicentre observational retrospective study, we analysed a series of patients who presented with FM associated with BD. All patients were identified through the French Society of Internal Medicine Network. The BD diagnostic criteria were those of the international study group for BD, [2] and the FM diagnosis was defined by an FM simple score (FMSS) ≥2. [6] Control group: FM not associated with BD (BD À FM+ group) The control group was composed of patients with FM (BD À FM+) identified through retrospective analysis of FM diagnosed in the Lyon

Key points
• Focal myositis (FM) occurring during Behçet's disease (BD) appears to be part of the systemic vasculitis process.
• FM occurring during BD presents with a particular clinico-histological pattern.
• This work highlights the need for research on BD diagnostic criteria in cases of FM.
University Hospital over a 19-year period. The eligibility criteria were an FMSS ≥2, without associated BD or known inflammatory disorders.

Clinico-pathological data collection and analysis
For all cases, medical features, laboratory, radiological (magnetic resonance imaging [MRI]) and electrophysiological data, as well as treatment and outcome were recorded. Patients were eligible if they met the inclusion criteria and had adequate data collection and material for the intended investigations. This study was conducted in accordance with the Declaration of Helsinki. All patients described in this study gave written informed consent. The study protocol has been validated by the local ethics committee (CRB-HCL Hospices Civils de Lyon BB-0033-00046).

Statistical analyses
Clinical, demographic, and laboratory data were analysed, and descriptive statistics were computed for each variable: numbers and percentages for categorical variables, medians and interquartile ranges [IQRs] for continuous variables. Comparisons between groups were made using nonparametric tests (Wilcoxon test and Fisher's exact test for quantitative and qualitative variables, respectively). Differences with a value of p < 5% were considered significant. In cases of significance, nonparametric post hoc tests were used. Analyses were performed using R software (version 3.6.0, R Computing, Vienna, Austria, 2019).
HLAB51 was positive in 5/8 tested patients. Individual patients are presented in Table 1.

Characteristics of the FM
FM was unifocal in seven cases and multi-monofocal (involving a maximum of three muscles) within the same or different compartment in three cases ( Figure 1). The anatomical distribution was mainly the lower limb (n = 6), followed by the orbital area (n = 3) and upper limb (n = 1). The gastrocnemius was the most frequently involved muscle (n = 4). The main muscular symptom was focal pain (n = 10), followed by local inflammatory symptoms (cutaneous redness and warmth; n = 8); no muscle weakness was reported. Among the three cases of orbital myositis, diplopia and proptosis were present in two patients. Creatine kinase levels were normal (n = 8) or mildly elevated (maximum two-fold elevation; n = 2). Electrophysiological investigations were abnormal in all tested cases, indicating focal myogenic features. MRI found focal inflammation in all patients ( Figure 1). Muscle biopsies were performed in six cases. Inflammatory infiltrates were observed in all cases; these were perivascular (n = 6), perimysial (n = 3) and/or endomysial (n = 2; Figure 1)  Table 2).

Characteristics of a BD À FM+ group and comparison with the BD + FM+ group
A total of 35 patients were included in the BD À FM+ group.
There was a trend towards a younger median age at diagnosis of  Figure S1).

DISCUSSION
In the present study, we found that patients with BD and FM differ from those with only FM on several points through an exhaustive analysis of the clinical presentation, imaging, histopathology, treatment, and outcome features.
The association of FM and BD appears rare but is probably underdiagnosed because of the varying clinical picture of BD, which can involve numerous organs. The median age at diagnosis of FM in those without BD was slightly higher than reported in the literature (ranging from 41-44 years), which may be explained by the inclusion in published FM series of potential BD cases or autoimmune-associated FM for which the age of FM diagnosis is younger. [3,5,6] Table S1). [2,6] Regarding the chronology of the development of the diseases, BD seems to occur first; herein, BD diagnosis preceded FM in all-but-two cases, which is consistent with the literature apart from a recently published study in which FM was considered as the primary manifestation leading to the diagnosis of BD. [13] However, the appearance of BD clinical diagnostic criteria relative to FM occurrence was not assessed. [13] Interestingly, FM onset was associated with concomitant BD flare-ups in some cases.
This is consistent with that reported in the literature (14/32 published cases; Table S1).
The location of FM in the cases of BD was similar in the present study to that reported in the literature (upper limb: 9%; lower limb: 69%; orbital: 19%; upper and lower limb: 3%), and FM had a severe clinical presentation, notably relating to focal oedema and pain (Table S1). However, as usually described for FM, [3] BD + FM+ patients did not present with weakness or functional impairment apart from pain-induced motionless and proptosis identified in cases of orbital myositis. This is consistent with the normal creatine kinase level and the relatively undamaged myofibres on pathological analysis of BD + FM+ patients. Interestingly, no association with cancer was found in contrast to BD À FM+ patients. It is worth noting that no FM evolved to generalised myositis, nor were they associated with myocarditis in our cases or those in the literature (Table S1).
FM associated with BD seems to run a relapsing-remitting course for which medical management was complex, often requiring a second or third line of immunosuppressive drugs in the cases presented herein; this is in sharp contrast to the favourable course found in FM. [6] However, no consensus exists for the care of peripheral ner- Regarding other vasculitides, a recent systematic review identified 395 cases of vasculitides associated with myositis in the literature. [15] Despite the incomplete characterisation (pathological analysis sometimes and whole-body MRI often lacking), most cases did not present with the usual clinical features of diffuse myositis; in contrast, isolated muscles or a combination of two or three muscles with an asymmetrical distribution were reported, notably involving the posterior compartment of the lower limbs. In large vessel vasculitis (Giant cell arteritis, Takayasu's arteritis), the authors discuss the possibility of the expression of a primary inflammation of muscles or secondary spread of vasculitis from muscle vessels, while in medium vessel vasculitis (mainly polyarteritis nodosa) myositis is found. Regarding small vessel vasculitis, the published data mainly concern anti-neutrophil cytoplasmic antibody-associated vasculitis, and the most frequent of these is microscopic polyangiitis. Among the studies included in the systematic review, the paper published by Lacou et al. [16] is of note; in a group of 78 patients with anti-neutrophil cytoplasmic antibodyassociated vasculitis who underwent systematic muscle biopsy, 58% of patients had a positive biopsy (with the presence of inflammatory infiltrates in vessel walls, with or without fibrinoid necrosis). While supported by an asymmetrical distribution and normal creatine kinase, the focal character of the myositis was not investigated or demonstrated for these patients and is questioned by diffuse myalgia and positivity of nonguided quadriceps biopsy. Taken together, these findings support the possibility that muscle is a target organ of vasculitis and elicit discussion regarding the occurrence of either fortuitous concomitant myositis or secondary vasculitis-related myopathy. This question requires additional more detailed investigations. In the present study, the integrity of myofibres and the absence of major modification of the connective tissue domain suggest that vasculitisrelated myopathy seems to take place focally in BD as is the case for other organs, such as unilateral uveitis, asymmetric aphtosis or skin lesions. Furthermore, diffuse sarcolemmal MHC-I overexpression and endomysial inflammatory infiltrates in rare cases could be explained by histological disturbances, secondary to vasculitis. These aspects support the hypothesis that FM has to be considered as a symptom of an acquired vasculitis-associated focal myopathy.
The present study does have some limitations. For instance, the number of patients was limited, and the study was retrospective. Nevertheless, the very low prevalence of each clinical entity, 7/100,000 for BD and 1.8/100,000 for FM, [1,3] makes the case series design the only feasible approach. Moreover, with a long latent period between diagnosis of BD and occurrence of FM (a median interval of several years), a prospective study remains difficult to conduct.
Despite these limitations, the study provides a detailed overview of this association.
In conclusion, FM occurring during BD appears to be part of the systemic vasculitis process and presents as a vasculitis-associated focal myopathy with a specific clinic-histological pattern. Patients with this association require long-term follow-up and adapted management. This case series also highlights the need for research on diagnostic criteria in the case of FM occurring in BD.

AUTHOR CONTRIBUTIONS
Substantial contributions to the study conception and design were