To the Editor:

Here we report the case of a patient with Crohn's disease (CD) characterized by a severe course and the occurrence of multiple extraintestinal manifestations, some of them well-known and defined, others rarely described in the literature.


The patient is a 41-year-old man who came to our attention in October 2009, presenting a long history of ileocolic steroid-dependent CD, starting with fever, perianal fistulization, and rectal bleeding 20 years before (1989). We present a summary of his clinical history.

Since 1993 the patient experienced episodes of severe exacerbations accompanied by extraintestinal relapsing manifestations: one-sided asymmetrical seronegative arthritis, pyoderma gangrenosum, and erythema nodosum. He was first treated with sulfasalazine (SASP), local and systemic steroids, then with methotrexate, 6-mercaptopurine, without achieving full remission. More than once surgical curettage of perianal disease and local infusion of infliximab were needed. In 1998 he underwent ileocecal resection for an obstructive complication and concomitant abdominal abscess with fistulization.

A month later a severe aphthous stomatitis with cervical lymph nodes inflammation and fever occurred: bacterial culture on various specimens (blood, excretion, swab) and serological tests (for cytomegalovirus [CMV], human immunodeficiency virus [HIV], Epstein–Barr virus [EBV], toxoplasma, Echo-Cox virus) excluded the most probable infective etiologies; this appearance was first considered an extraintestinal localization of CD—it responded to high doses of steroids, but relapsed at steroid tapering, becoming a chronic lymphadenitis, with a minimal effective dose of methylprednisolone 16 mg. The analysis of lymphocyte subpopulations did not demonstrate any abnormality. Chest x-ray, abdominal ultrasound, echocardiography, skin test for anergy, and ACE were normal. Finally, alcohol-acid-resistant bacilli were found in the expectorated analysis and the histological lymph node examination revealed the presence of a chronic granulomatous necrotizing suppurative lymphadenitis probably due to an atypical mycobacteriosis, which seemed to respond to a 4-fold antimycobacterial therapy for addiction to a tapering course of steroids. However, the same appearance, including an intermitting fever, relapsed 3 months later (1999); at this time the research for BK or atypical mycobacteria on biopsies was negative and transesophageal echocardiography, bone marrow biopsy, and bronchoscopy did not suggest anything significant. The treatment with 6-mercaptopurine failed due to intolerance, while cyclosporine was ineffective; on the contrary, the patient experienced two episodes of seizures and a left fronto-rolandic intracerebral deep lesion was diagnosed, requiring a neurosurgical intervention.

In 2003 he had a relapse of lymphadenitis with evidence of splenomegaly associated with platelets depletion; the analysis of osteomedullary biopsy showed no abnormality.

From 2006 to 2009 the progressive increase of the spleen size, monitored through ultrasound every 6 years, led both the hematologists and the gastroenterologists to consider a hypothesis of splenectomy.

When he came to our attention (October 2009) he reported a good well-being when taking a daily dose of 5 mg beclometasone, which controlled the fever. He did not present any sign of lymphadenitis. He described an increase of temperature following every attempt to discontinue steroids. After performing the necessary screening tests, we decided to introduce a biologic therapy with adalimumab following the classic administration schedule, thus managing to reduce the beclometasone dosage (5 mg every 3 days) within 2 months. Inflammatory markers remained elevated. Then the tomographic evidence of marked splenomegaly (bipolar diameter of about 20 cm) with splenic vessels ectasia, left kidney compression and dislocation, and multiple enlarged abdominal lymph nodes caused us to consider a new surgical and hematological consultation, which led to the resolution of splenectomy in January 2010 (surgical specimen size: 19 × 18 × 12 cm). The histological examination revealed a congestive microgranulomatous splenomegaly etiologically attributable to CD. The granulomas, localized in some follicles within the white pulp and also in splenic perihilar lymph nodes, were noncaseating, epithelioid granulomas without necrosis, containing neutrophil infiltration rarely showing aseptic microabscesses (Fig. 1). During the following months the patient condition improved, although he was not steroid-free yet. Unfortunately, a few months later (July) an acute myeloproliferative syndrome occurred (cytotype FAB M1) which needed a polychemotherapy treatment. Our patient quickly gained remission after the induction cycle, but the leukemia phenotype requires bone marrow transplantation, for which the patient is still waiting for a donor. Concerning the fever and the inflammatory disease, he gained a complete remission after chemotherapy treatment.

thumbnail image

Figure 1. (A,C) Single nodule-epithelioid granuloma within the splenic parenchyma. (B) Confluent granulomas within the splenic parenchyma. (D) Microgranulomas within a splenic perihilar lymph node. [Color figure can be viewed in the online issue, which is available at]

Download figure to PowerPoint


In CD, the same kind of granulomas observed in the gut has been found in a wide variety of organs, in or extraabdominal. The association between IBD and neutrophilic dermatosis (Sweet's syndrome or pyoderma gangrenosum), characterized by sterile collection of polymorphonuclear neutrophils, is well established.1 However, the finding of splenic granulomas is anecdotal. To our knowledge, three cases have been reported in the literature. The first was the case of a 42-year-old patient with a 10-year history of CD who had presented with hepatosplenomegaly, fever, and left upper quadrant pain. He was found to have noncaseating granulomas of the liver, spleen, and bone marrow which promptly and fully responded to high-dose prednisolone therapy.2 The second was a 34-year-old man with a recent diagnosis of inflammatory bowel disease who underwent splenectomy, as he had presented with severe left upper quadrant pain with laparotomic findings of an enlarged spleen and a small amount of free blood in the peritoneal cavity. The microscopic examination of the spleen showed marked sinusoidal congestion and a single granuloma containing a small number of neutrophils without necrosis.3 In the third case, a 52-year-old man with known CD and primary sclerosing cholangitis (PSC) in clinical remission showed multiple lesions of the spleen in routine MR tomography. The histological examination showed noncaseating epithelioid cell granulomas, which regressed completely and spontaneously without specific immunomodulatory therapy.4 In general, the involvement of the spleen is rare in CD; Coyne3 suggested a correlation between spleen granulomas and disseminated sterile abscesses characterized by a granulomatous pattern, described as a well-defined clinical-pathological entity in some cases of CD from 1995, i.e., the aseptic abscess syndrome (AA).3, 5 The symptoms and abscesses do not respond to antibiotics, but improve quickly with systemic corticosteroids or other immunosuppressive therapies.

In our case, among the various manifestations, the patient had first a history of relapsing fever and lymphadenopathy and then a long-lasting history of splenomegaly. The histological examination of cervical lymph nodes and spleen revealed different pattern of lesions, respectively a chronic granulomatous necrotizing suppurative lymphadenitis and microscopic noncaseating, epithelioid granulomas without necrosis. In the first case a mycobacterial etiology was supposed following the finding of alcohol-acid-resistant bacilli at the expectorated analysis; however, this appearance did not fully respond to the antimycobacterial therapy and, after relapsing, no bacteria could be found. Therefore, this appearance was finally considered an extraintestinal manifestation of CD. Concerning the splenic lesions, they did not show macroscopic abscess features, but they were comparable to typical Crohn's granulomas. Splenomegaly was probably due to sinusoidal congestion and expansion. Neither steroids nor biologic therapy could control the increase of spleen size (only partially the fever); splenectomy was unavoidable and only the histological examination allowed us to formulate an etiological hypothesis for splenomegaly in the light of the clinical picture in a patient with such a complex pattern of inflammatory disease: all our findings suggest an extraintestinal localization of CD. Finally, after splenectomy our patient faced an acute myeloproliferative syndrome. Is it a coincidence? We cannot answer this question, as we cannot determine with certainty how much the immunosuppressive therapy facilitated this occurrence.


  1. Top of page
  • 1
    Kemmett D, Gawkrodger DJ, Wilson G, et al. Sweet's syndrome in Crohn's disease. Br Med J. 1988; 297: 15131514.
  • 2
    Puli SR, Presti ME, Alpert MA. Splenic granulomas in Crohn disease. Am J Med Sci. 2003; 326: 141144.
  • 3
    Coyne JD. Crohn's disease with inflammatory splenic granulomas. J Clin Pathol. 2006; 59: 889.
  • 4
    Schulz C, Von Arnim U, Kuester D, et al. Splenic granulomas — rare extraintestinal manifestation of Crohn's disease. Z Gastroenterol. 2010; 48: 13671370.
  • 5
    André M, Aumaître O, Marcheix JC, et al. Aseptic systemic abscesses preceding diagnosis of Crohn's disease by three years. Dig Dis Sci. 2010; 40: 525527.

Lucia Castellani MD*, Alessandro Sartini MD*, Margherita Marocchi MD*, Tiziana Balbi MD†, Andrea Belluzzi MD*, * Department of Internal Medicine and Gastroenterology, S. Orsola-Malpighi Hospital, University of Bologna, Bologna, Italy, † Department of Haematology Oncology and Laboratory Medicine, University of Bologna, S. Orsola-Malpighi Hospital, Bologna, Italy.