Gastrointestinally mediated food allergy causing Spondyloarthritis-like disease

A 56-year-old woman with seronegative spondyloarthritis observed arthritic fl ares following ingestion of certain foods. Blood analyses were unremarkable. Segmental gut lavage revealed plasmacellular infi ltrates and eosinophilia in terminal ileum. Food allergen-specifi c IgE-analysis of lavage fl uid showed signifi cant polyvalent intestinal IgE-sensitization (>0.35kU/mg protein) towards wheat, rye, egg, soybean, pork, beef, nuts. Repeated exposure to the aforementioned foods caused arthritis within 48 hours; elimination diet engendered long-term remission. Seronegative local IgE-mediated GMA (type I) was diagnosed.

Although patients with allergy sometimes report arthralgia or arthritis, its occurrence has only been observed rarely in Double-Blind, Placebo-Controlled Food Challenge tests (DBPCFC) or in case reports [4,5].
Hvatum, et al. [9], explored the role of the intestine in Rheumatoid Arthritis (RA) and found increased IgG, IgA, and IgM antibodies to dietary antigens more frequently in jejunal perfusion fl uid than in serum of 14 seropositive RA patients. There is a growing awareness that in infl ammatory autoimmune diseases such as RA or Spondyloarthropathy (SPA) certain patients may benefi t from diet change, either because of epigenetic infl uences of nutrition on chronic infl ammation, content of proinfl ammatory mediators in certain foods (e.g. arachidonic acid), indirect or direct effects on gut microfl ora and/or intestinal immune cell activation [5,9]. Hence, an unknown number of RA-or SPA-patients might suffer from leaky gut syndrome, unapparent GMA, nonceliac gluten sensitivity or cross-reactions of food antibodies to immunoglobulins triggering infl ammation and arthritis.

Case report
We present a 56-year-old female patient earlier diagnosed with seronegative peripheral Spondyloarthritis (pSPA). Pain and swelling of shoulders, hips, knees and upper ankle joints fi rst appeared in 2008. During hiking the patient tried therapeutic fasting, but due to fear of loss of muscle strength she added buttermilk and brown bread to her diet of vegetable juices. Her arthritis exacerbated quickly; thus, she consulted a rheumatologist who suspected seronegative pSPA. Non-Steroidal Anti-Infl ammatory drugs (NSAIDs) relieved the symptoms. In the sequel, oligoarthritis occurred sporadically. She then eliminated milk-and gluten-containing products from her diet, which improved the arthritis. Exposures to the respective comestibles reproducibly caused exacerbation of arthritis within 48 hours; therapeutic fasting (vegetable juices only) completely terminated the joint complaints. The patient fi rst did not allocate her co-occurring gastrointestinal symptoms (bloating, pain, slight diarrhea) to foods. A then unknown trigger gradually caused arthralgia in both temporomandibular and all metatarsophalangeal joints; arthritis in previously affected areas persisted. The patient presented at our gastroenterologic and rheumatologic clinics because of the observed gastrointestinal symptoms.

Clinical, laboratory and endoscopic testing
Clinical examination gave slight joint pain of knees and upper ankle joints without effusion. Morning stiffness was 5 minutes. Blood analyses for C-reactive protein, blood eosinophils, histamine, serum-ECP, tryptase, TNF-alpha, total and food-specifi c IgE in serum as well as rheumatology diagnostics including rheumatoid factor, anti-CCP-antibodies, anti-mutated citrullinated vimentin, anti nuclear antibodies and interleukin-6 were unremarkable.
Before performing ileo-colonoscopy for local IgE detection from lavage fl uid, further evidence for ongoing intestinal allergic reaction was obtained from elevated urinary methylhistamine excretion. During an unrestricted diet with ingestion of all staple foods Urinary Methylhistamine (UM) levels were clearly elevated (9.2μg/mmol creatinine×m 2 body surface area (BSA); normal <6.5), while during an oligoantigenic elimination diet (2 days) with potato, rice and water UM decreased to 6.2μg/ mmol creatinine×m 2 BSA [3,10].

Endoscopically guided segmental gut lavage
Abdominal ultrasound, upper and lower endoscopy gave no pathological results. Therefore, an endoscopically guided segmental lavage including biopsies was performed during ileo-colonoscopy as described previously [2,3,11]. Briefl y after insertion of the endoscope at ileum, cecum, and rectum, 50ml of saline solution were installed in the gut lumen for one minute. Afterwards, at least 10-15ml of the fl uid was suctioned into containers holding protease inhibitors (0.1mM EDTA, AEBSF-HCL, Pefablock 0.5mM and 420nM Aprotinin). The cooled lavage fl uid was aliquoted for detection of Eosinophilic Cationic Protein (ECP), tryptase and TNFa by fl uoro-enzyme immunoassay (Cap-FEIA, Thermo Fischer, Freiburg Germany) and ELISA (IBL, Hamburg, Germany), respectively [2,11]. For detection of total IgE and food-specifi c IgE the lavage fl uid was centrifuged (4000×g), ultrafi ltrated and 10-fold concentrated (Vivaspin20, Sartorius, Germany). The concentrated lavage fl uid was then dissolved with sample IgE diluent (Thermo Fischer, Freiburg, Germany) and analyzed with ImmunoCAP 250 (Thermo Fischer, Freiburg, Germany) for total and foodspecifi c intestinal IgE using the high-sensitive IgE standard [2,3,11]. Mediators and IgE levels were expressed in relation to the protein content of the lavage fl uid. Food-specifi c IgE was considered positive when IgE levels were >0.35 U/mg protein at one lavage site [11].
As previously published, healthy controls did not have food-specifi c IgE antibodies in their gut lavage fl uids [2,11].

Histopathology and intestinal immunological results
Endoscopically guided segmental lavage detected a sitespecifi c, signifi cantly enhanced local IgE production in the terminal ileum (Table 1), but not in the cecum or rectum. Highest local IgE titers in our patient were found in response to wheat and rye, but also soy, pork, beef, egg and nuts showed elevated concentrations.
Interestingly, comparison between immune mediators and IgE from blood and ileum fl uid gave evidence of a pathological immune response clearly restricted to the small bowel compartment terminal ileum. Herein, there was a slight accumulation of lymphoid follicles, plasmacellular and eosinophilic infi ltrates with moderately increased mucosal mast cell numbers; expression of diaminooxidase was low (Figures 1a-c).

Confi rmation of local gastrointestinally mediated allergy (GMA) by food challenge tests
Open standardized food challenge tests with wheat, rye or meat reproducibly exacerbated arthritis, meteorism, moderate Citation: Raithel  On a milk-and gluten-free diet pain on a visual analogue scale (VAS; 0-100mm) was 60mm, rose to 80mm after exposure to all food allergens, and fell to 20mm undergoing allergenfree diet. VAS for patient global health was 30mm, 50mm and 10mm; BASDAI was 3.2, 6.8 and 1.2, respectively.
The patient was thereby diagnosed as seronegative local type I GMA.

Discussion
Food allergy may involve intestinal and extraintestinal organs, especially known in atopic individuals as Th2mediated disease. While this type of FA is often characterized by evidence of sensitization to food antigens in blood or skin, GMA may present as seronegative allergic disease, which is diffi cult to recognize unless provocation tests or specialized immunological tests at intestinal level are applied [2,3,[10][11][12][13][14]15]. Presence of local IgE within mucosal surfaces without systemic IgE elevation has recently been introduced as entopy and found to induce particularly local or functional symptoms [14,15].
Interestingly, our patient with six years of recurring oligoarthritis fulfi lled criteria of local gastrointestinal IgE sensitization to food antigens with classical criteria of entopy (bloating, irritable bowel, pain, diarrhea). However, peripheral symptoms outside the entopic organ have not yet been described in patients with local IgE production, and at present it is unclear how this local immunopathology contributed to induction of musculoskeletal symptoms, which was confi rmed at two further food challenge procedures. Since avoidance of these food antigens reproducibly led to complete remission of arthritis, we are able to describe an as yet unknown mechanism connecting gut mucosal IgE pathology with the musculoskeletal system. As yet, data on this issue are extremely scarce, despite some patients reporting food-induced musculoskeletal symptoms, but such connections have not been proven in detail in terms of immune-mediated FA or non-immune food intolerance except in rare cases [4][5][6]8,9].
Ingestion of intestinal IgE-positive food antigens resulted in crosslinking of intestinal immune effector cells like mast cells or eosinophils. Their mediators within the GIT may be responsible for peripheral organ symptomatology, but it is also possible that absorbed food antigens reach the synovia and induce swelling, effusions or pain when targeting synovial or connective tissue mast cells. However, the clinical course of symptoms does not support these theories, as musculoskeletal Table 1: Comparison of food-specifi c IgE fi ndings in blood and from intestinal lavage fl uid of the terminal ileum in the patient with recurrent food-dependent oligoarthritis. Endoscopically guided segmental lavage was performed at ileum, caecum and recto-sigmoid according to the methods published previously [2,3,11]. The same food antigens were used by CAP-FEIA (ThermoFischer, Phadia, Freiburg, Germany) to detect food-specifi c IgE in serum (n < 0.35 kU/L) and at the intestinal level in ileum (intestinal IgE value n < 0.35 U/mg protein). Intestinal IgE in caecum and rectosigmoid was negative, indicating local allergic enteritis of the ileum. Values marked by asterisk indicate pathological fi ndings. IgE: Immunoglobuline E.  Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Food
Baenkler HW: Analysis, and interpretation of data for the work; Revising the work critically for important intellectual content; Final approval of the version to be published.
Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved