POS1336 RETROPERITONEAL FIBROSIS IN ERDHEIM-CHESTER DISEASE HAS UNIQUE PRESENTING AND PROGNOSTIC FEATURES: A SINGLE CENTRE RETROSPECTIVE COMPARATIVE COHORT STUDY

Abstract

Background:Erdheim-Chester disease (ECD) is a multisystemic non-Langerhans cell histiocytosis. Retroperitoneal fibrosis (RPF) is a common ECD feature, found in up to 60% of patients. Besides ECD, RPF differential diagnosis includes other clinical entities of rheumatological interest, such as idiopathic RPF and IgG4-related disease (IgG4-RD). Since ECD management does not include glucocorticoids and mainly relies on the use of anti-neoplastic small molecules, a prompt diagnosis is critical.Objectives:To characterize the main differential clinical, radiologic and prognostic features of a monocentric cohort of patients with RPF. To provide practical learning points useful to recognize ECD among patients with RPF.Methods:Medical records of ECD patients with radiologic signs of RPF and of patients with non ECD-related RPF followed up at our Rheumatology Clinic were retrospectively reviewed. Final diagnosis relied either on analysis of histologic samples or on fulfilment of appropriate classification criteria. Main clinical, radiologic and laboratory features at diagnosis and long-term outcomes were evaluated and compared between the two groups. Non parametric tests were used for comparison.Results:31 (41%) ECD and 45 (59%) non-ECD patients (IgG4-RD, n=28; idiopathic RPF, n=11; peri-aortitis with large-vessel involvement, n=6) were included. All but one (97%) ECD patients and 56% non-ECD patients had a biopsy-proven diagnosis. ECD patients were younger at disease onset (51±14 vs 59±10 years, P=0.024) and diagnosed with a greater delay (48 [IQR, 13-92] vs 4 [IQR, 2-7] months, P<0.001). In the majority of ECD patients (84%), RPF was clinically silent and was disclosed in the context of the diagnostic work-up; conversely, 82% non-ECD patients had at least one RPF-related symptom (mainly, abdominal pain, n=19, and low-back pain, n=16). At diagnosis, C-reactive protein (50±34 vs 33±31 mg/L, P=0.052) and serum creatinine (1.31±0.53 vs 1.34±0.79, P=0.76) levels were not statistically different. In all ECD patients, fibrotic tissue surrounded both kidneys symmetrically, involving the ureters in half of the cases (Table 1); signs of hydronephrosis were present in 55% of patients (mostly bilaterally, 77%). In the non-ECD group, kidneys were rarely affected, and ureteral involvement was mostly asymmetric; hence, hydronephrosis incidence was similar (76%), but more frequently monolateral (65%). In non-ECD patients, fibrotic tissue more frequently involved inferior vena cava and iliac arteries. FDG uptake by RPF was disclosed in 39/40 (98%) non-ECD and 5/12 (42%) ECD patients who underwent a PET before therapy start (P<0.001). Ureteral stenting was required by a similar fraction of patients (ECD, 29%; non-ECD, 38%, P=0.47), but it was bilateral in 100% of ECD vs 29% of non-ECD patients (P<0.001). No differences in the rate of chronic kidney failure was found (26% vs 18%, P=0.41), but only ECD patients progressed to a terminal stage requiring hemodialysis (13% vs 0%, P=0.025).Table 1.ECD patients (n=31)Non-ECD patients (n=45)P-valueLeft kidney30 (97%)4 (9%)<0.001Right kidney31 (100%)6 (13%)<0.001Left ureter16 (52%)16 (36%)0.24Right ureter16 (52%)24 (53%)1Abdominal aorta21 (68%)40 (89%)0.04Inferior vena cava5 (16%)29 (64%)<0.001Left iliac artery10 (32%)27 (60%)0.02Right iliac artery10 (32%)29 (64%)0.001Left renal artery9 (29%)6 (13%)0.14Right renal artery6 (19%)4 (9%)0.3Mesenteric artery8 (26%)4 (9%)0.06Celiac trunk7 (23%)3 (7%)0.08Conclusion:Our retrospective study suggests that ECD-related RPF has a significantly worse renal prognosis than non ECD-related RPF. A complete baseline work-up including a critical evaluation of the extension of the fibrotic tissue and the sites involved is fundamental to avoid patients’ misdiagnosis and mistreatment.