The pathogenesis of CMLNS is not fully understood at the moment. One of the proposed theories involved an excessive antigenic exposure of the immune system via a damaged intestinal mucosa, leading to depletion of cellular lymphoid elements in the mesenteric lymph nodes and spleen, causing a cystic change or ‘cavitation’ in some patients with celiac disease. An alternative hypothesis is the necrosis of mesenteric lymph nodes triggered by localized immune-mediated complement activation and intravascular coagulation . There were several reports of lymph node necrosis associated with intestinal infection with Mycobacterium spp, Yersinia spp  or Tropheryma whippelii . Our patient had an infection with Yersinia enterocolitica during the course of his illness. However, this infection cannot be regarded as the cause of CMLNS, because the symptoms persisted even after adequate antibiotic treatment leading to stool sterilization.
Half of the CMLNS patients have a poor prognosis, but some reports mentioned a good outcome if a strict gluten-free diet was observed [10, 11]. In celiac disease, however, a malignant lymphoma may also be the cause of mesenteric lymph node necrosis, being in part responsible for the poor prognosis .
The diagnosed of CMLNS is based on imaging and typical histopathological examination. The necrotizing mesenteric lymph nodes are described as anechoic cysts on an abdominal ultrasound. Computed tomography studies reveal central low attenuation with enhancing rims . Sometimes fat-fluid levels within the masses may be apparent on CT ; this feature is considered unique to the cavitated mesenteric adenopathies associated with celiac disease. The specific fat-fluid levels were also found in one CMLNS case on an MRI examination .
The diagnosis of complicated cavitating mesenteric lymph node syndrome (either infectious or malignant) can be challenging; a lymphoma is still notoriously difficult to diagnose, despite multiple biopsies . Lymphomas with T-cell immunophenotypic features have been reported in extra intestinal sites (liver, spleen), as a complication of celiac disease, but without intestinal involvement . In our case, the initial biopsies did not reveal abnormal T-cell lymphocytes, but the virtual lack of immunopositivity for CD8 intraepitelial lymphocytes might have been suggestive of a diagnosis of refractory celiac disease type II . Also the presence of isolated large CD30+ lymphocytes in the lamina propria in early biopsies could have represented minimal infiltration by the patient’s lymphoma .
The capsule endoscopy and CT did not find any small bowel mucosal changes suggestive of enteral lymphoma and no other small bowel biopsies were taken. The contrast-enhanced ultrasound was performed in order to obtain more specific information about the vascular pattern of the lymph node “walls” and the surrounding mesenteric tissue.
The assessment of neovascularization by CEUS is based on its ability to depict the blood flow in small vessels. In malignant tumors a rapid and intense enhancement is seen in the arterial phase, with rapid wash-out of the contrast agent in the venous phase. This vascular pattern is explained by arteriovenous shunts. The timing of hypo enhancing on CEUS may be correlated with tumor cell differentiation; well-differentiated tumors wash out more slowly than poorly differentiated ones . The quantitative CEUS parameters in assessing neoangiogenesis processes were documented on hepatocarcinoma, ovarian and breast malignant tumors [15–17]. In malignant lymph nodes, CEUS depicts vessels penetrating the node’s capsule away from the hilum; reactive adenopathies have a singular vascular pedicle at the hilum with regular branches towards the periphery. Based on these findings, CEUS can improve the results of Doppler ultrasound in the differential diagnosis of lymph nodes with a sensitivity, specificity and accuracy rate of up to 84%, 79% and 80% respectively . These studies were performed on superficial lymph nodes. However, in several studies, lymphomas seem to have a benign vascular pattern [19, 20]. To date there is not enough evidence for the accuracy of CEUS in assessing different types of lymphoma . In our case, intense enhancement was observed in the arterial phase in the surrounding mesenteric tissue of necrotic lymph nodes, without rapid vascular washout in the venous phase, but it was not suggestive of malignancy. CEUS failed to provide a diagnosis of tumor neoangiogenesis in this case. This lack of venous wash-out may be due to arborizing venules found in some peripheral T-cell lymphomas .
In this case, 18 F-FDG PET scan could have detected early EATL; previous studies documented its role in patients with refractory celiac disease, being more sensitive than CT in detecting sites affected by lymphoma . This investigation was not available in our institution.
In patients associating CMLNS, liver failure may be due to septicemia or to malignant T-lymphocyte infiltration of the liver [2, 3]. In our case, the microscopic examination revealed abnormal T-cell lymphocytes in the liver, spleen, mesenteric tissue, gastric walls, kidney, lung and bone marrow. No malignant cells were observed in the small bowel samples examined. It is possible that the EATL was unsampled, despite multiple biopsies, as it was previous reported . Other authors described the same features of a rare peripheral T-cell lymphoma, associated with celiac disease, characterized by a rearrangement of the gamma-delta T-cell receptor, responsible for the aggressiveness of this tumor [2, 24]. T-cell receptor PCR or flow cytometry were not performed in our case.