Our patient presents the very uncommon association of thyroid, tracheal and nodal RDD with multiple small lung cysts. Only few cases of lung cysts associated with RDD are described in the literature. In fact, a recent Mayo case series of 9 patients with intrathoracic manifestations of RDD mainly found mediastinal lymph node involvement. However, there were two patients (#4 and #9) with lung cysts and bronchiectasis [
6]. Moreover, cystic lung disease has developed over time in a patient presenting with consolidative nodules, pleural thickening, septal lines, and mediastinal lymphadenopathy. The diagnosis of pulmonary RDD was established by open-lung biopsy [
8]. Another patient with cutaneous RDD showed bilateral pulmonary cysts and numerous ground glass nodules with characteristic upper lobe predominance. There was no intrathoracic lymphadenopathy or pleural effusion mimicking PLCH. However, the biopsy of the concomitant skin lesion confirmed the histological diagnosis of RDD [
9]. Thus, cyst-formation is possible in pulmonary RDD, it is however very rare and likely to be accompanied by other pulmonary radiographic findings, suggestive of interstitial lung disease. The main differential diagnosis of pulmonary cysts included PLCH, lymphocytic interstitial pneumonia (LIP), or lymphangioleiomyomatosis (LAM) [
10,
11]. In fact, histiocytosis (e.g. PLCH and Erdheim-Chester disease) frequently cause cystic lung disease with a mid to upper zone predominance [
12]. Nevertheless, the absence of tobacco consumption, the diffuse cysts localisation and the absence of peribronchiolar nodular opacities all argued against PLCH. Moreover, the patient did not present any of the typical features of Erdheim-Chester disease. The clinical presentation as well as the absence of pulmonary nodules and ground glass opacity was atypical for LIP. Furthermore, human immunodeficiency virus (HIV) and immunological tests (anti-Ro/SSA and anti-La/SSB antibodies, rheumatoid factor and TRAb) were negative. Genetic testing for Birt-Hogg-Dube syndrome (BHDS) was not performed since there was no family history, no extrapulmonary manifestations, such as skin lesions or renal cancer, and the distribution of the cysts was not suggestive. Lung cysts in BHDS are multiple, thin walled, typically seen in the peripheral zones at lung bases and along the mediastinum with a disproportionate number of paramediastinal elliptical (floppy) cysts, and can abut or encase the proximal portion of the lower pulmonary veins [
11]. In contrast, the multiple small and thin-walled lung cysts in the presented case were higher in number and diffusely distributed throughout the lung parenchyma, more ressembling the distribution in LAM. The radiologic appearance of the cysts, age and sex of the patient were indeed compatible with LAM. The VEGF-D plasma level was low (347 pg/mL) neither confirming nor excluding LAM since VEGF has a good positive predictive value if the value is greater than 600 pg/mL but a poor negative predictive value below this threshold [
13]. In addition, abdominal CT scanning did not show any sign of renal angiomyolipoma which is associated to LAM in 50% of the cases. Taken together, without lung biopsy the observed cystic lung lesions cannot with certainty be attributed to RDD, since a concomitant diagnosis of LAM or BHDS was not completely ruled out. Although two rare diseases may coexist in the same patient, it is preferable to retain a single etiology and unifying diagnosis. Therefore, in analogy to other histiocytosis known to cause lung cysts the pulmonary lesions in our case are probably related to RDD.
In general, associated autoimmune diseases can be found in 13% of RDD. As demonstrated in the our case a subset of patients with RDD present increased levels of IgG4-bearing plasma cells upon immunohistological examination rendering the differential diagnosis with IgG4-related disease somewhat difficult [
1,
14]. The overlap between RDD and IgG4-related disease with the two conditions sharing similar features, including emperipolesis, may represent a spectrum [
15,
16], whereas other authors suggest differentiation between the two based on the degree of IgG4-positive infiltrates and IgG4:IgG ratio [
17]. Menon et al. reported the presence of IgG4-bearing plasma cells in 28/70 RDD cases (40%) with 17.4% fulfilling the consensus guideline criteria for a histological diagnosis of IgG4-related disease [
18]. Of note, the two cases of intrathoracic RDD with elevated IgG4-bearing plasma cells presented with pulmonary masses and not with cysts. The presence of abundant emperipolesis was highly suggestive for RDD and the levels of IgG4-bearing plasma cells insufficient to make a diagnosis of IgG4-related disease. However, the significance of IgG4-bearing plasma cells in RDD remains to be further elucidated, and their presence might indicate a better treatment response to corticosteroids and/or rituximab in case of disease progression.
In conclusion, this case report indicates a benign course of nodal RDD with thyroid and tracheal infiltration following surgical resection, describes an association of typical histological signs of RDD (emperipolesis) with IgG4-related disease features, and that lung cysts might be a manifestation of RDD.