This previously healthy 4year-old girl was seen at an outside hospital for evaluation of fever and abdominal pain of a 2- week duration. There was no history of vomiting, diarrhea, rash, arthritis, or other symptoms. The family and past medical histories were unremarkable. There were no known exposures to animals, unpasteurized milk, uncooked meat, or tuberculosis.
On physical examination, she had a rectal temperature of 39.5° C, pulse 120, respiratory rate 20, blood pressure 100/60. Her growth parameters were in the 25th percentile. HEENT exam was unremarkable. There was no significant cervical adenopathy. Lungs were clear. Sl and S2 were normal; there were no murmurs, rubs, or gallops. Abdomen was soft and mildly distended with tenderness to palpation in both lower quadrants; there was no rebound tenderness or guarding. Rectal examination had Hemoccult® negative stool. Neurologic examination was unremarkable. On laboratory evaluation, the hemoglobin was 11.5 g/dL, white blood cell (WBC) count 17,600 with a differential count of 76% neutrophils, 14% lymphocytes, and 7% monocytes; platelet count 486,000. The erythrocyte sedimentation rate (ESR) was 70 mm per hour. A liver panel was normal. Computerized tomography (CT) of the abdomen and pelvis revealed multiple large mesenteric lymph nodes. There was no sign of appendicitis or bowel wall thickening. She was transferred for evaluation of a possible lymphoma. On review of the CT scan, neither the radiologists nor the oncologists felt that the mesenteric adenopathy resembled a neoplasm. A diagnosis of mesenteric adenitis was made, and she was discharged. Antibody titers to cat-scratch disease (B. henslae), adenovirus, toxoplasmosis, Epstein-Barr virus, and cytomegalovirus, as well as febrile agglutinins, were all negative. After another week of intermittent fever and abdominal pain, her symptoms resolved.
Robert Listernick, MD, general academic pediatrician: Is mesenteric adenitis a real entity?
Stan Shulman, MD, pediatric infectious disease specialist: The diagnosis of mesenteric adenitis is usually made in the operating room when a child, who has the clinical diagnosis of appendicitis, is found to have a normal appendix and large, inflamed mesenteric lymph nodes. Many of the pathogens that were just mentioned have been associated with mesenteric adenopathy. Yersinia enterocolitica is the classic pathogen that can cause the syndrome of fever and abdominal pain that mimicks appendicitis. In the underdeveloped world, calcified mesenteric lymph nodes due to Mycobacterium tuberculosis are a common finding.
Dr. Listernick: I've always thought that it was a bogus diagnosis made by surgeons to explain "appendicitis-like" abdominal pain.
Robert Arensman, MD, pediatric surgeon: In reality, it often is somewhat bogus. We find large lymph nodes in such children and the pathologist labels it "reactive adenitis." My impression has been that there is often an abundance of lymphatic tissue near the ileum in many small children who are being operated on even for reasons not related to fever and abdominal pain.
Grant Geissler, MD, pediatric surgeon: I agree, but there are definitely children who have right lower quadrant abdominal pain, fever, and adenopathy on CT scanning who don't have appendicitis.
Dr. Listernick: Can we see the original CT scan?
Cynthia Rigsby, MD, pediatric radiologist: There are multiple 1- to 1.5-cm lymph nodes in the left upper quadrant surrounding the mesentery and the mesenteric vessels. Although we see abdominal lymph nodes in most children who have abdominal CT scans, these are larger than most cases. The bowel wall appears normal.
Dr. Listernick: Is this the radiographic appearance of an abdominal lymphoma?
Dr. Rigsby: Burkitt's lymphoma is generally a single large mass, often in the right lower quadrant. In older patients, a lymphoma may invade the bowel wall leading to a hugely dilated thickened loop of bowel. These lymph nodes look neither large nor extensive enough to be concerning for a lymphoma.
Joanna Weinstein, MD, pediatric oncologist: Abdominal lymphomas are uncommon in so young a child. With that said, abdominal lymphomas often present with abdominal pain and fever mimicking appendicitis or as an asymptomatic mass. Lymphoma may be the lead point for an intussusception in an older child.
Dr. Listernick: Moving on, she recovered from this illness after a total course of 4 weeks. Six months later, she returned with abdominal pain and fever of 1-week duration. A repeat CT scan revealed impressively large mesenteric adenopathy to the same degree as on the previous study. In addition, there were mildly thickened duodenal and ileal bowel folds. She had grown and gained weight normally during this period. The only abnormal finding on physical examination was diffuse abdominal tenderness and voluntary guarding. There was no rebound tenderness. The WBC count was normal and the ESR was 50 mm per hour Serum albumen, antinuclear antibody, and a tuberculin test were all negative. Both IgG and IgA anti-gliadin antibodies were present; however, the tissue transglutaminase (TTG) antibody was negative. Upper and lower endoscopy of the gastrointestinal tract was normal. Her symptoms subsided after 2 weeks.
During the next 12 months, she had mild intermittent abdominal pain without any other symptoms. There was no discernible pattern to the pain. A CT scan during this period continued to show mesenteric adenopathy without any significant change. Three days prior to this admission, she developed fever, vomiting, and dehydration. On examination, she hadn't gained any weight during the preceding 12 months. The only finding on physical examination was diffuse abdominal tenderness with involuntary guarding. Rectal examination was unremarkable. The hemoglobin was 10.1 g/dL, WBC count 17,400 with 75% neutrophils, 16% lymphocytes, 6% monocytes; platelet count 545,000/mm3 and ESR 118 mm per hour.
Ben Katz, MD, pediatric infectious disease specialist: For what it's worth, we generally see sedimentation rates that high in inflammatory or rheumatologic diseases rather than with infections or malignancies.
Dr. Listernick: Let's backtrack a bit. How should we interpret the positive antigliadin antibodies?
B Li, MD, pediatric gastroenterologies The sensitivity and specificity of the IgG and IgA antigliadin antibodies are not very high. However, the TTG antibody has more than 95% sensitivity and specificity. Therefore, the likelihood that this child has celiac disease is low. One has to be careful in interpreting these tests because individuals who have both celiac disease and IgA deficiency, a relatively common disease affecting about 1% of the population, will have falsely negative titers.
Dr. Listernick: Could this child have inflammatory bowel disease (IBD)?
Dr. Li: IBD is certainly a possibility, although extensive mesenteric adenopathy is unusual and could even suggest an underlying intestinal lymphoma. There were also other subtle clues that spoke against IBD. Children with IBD generally have ESRs in the 50 to 60 mm per hour range, rather than 1 1 8 mm per hour, as in this child, unless they have a concomitant abdominal abscess. In addition, they usually have low-grade fevers rather than high fevers when febrile.
Dr. Listernick: What is the utility of the available antibody screening tests for IBD?
Dr. Li: That's controversial, even among pediatric gastroenterologists. These tests basically measure two sets of antibodies, the antineutrophilic cytoplasmic antibody (ANCA), which is associated with ulcerative colitis, and the anti-Saccharomyces cerevisiae antibody, which has been associated with Crohn's disease. The commercial lab has a two step screening process for detecting the presence or absence of these antibodies. If present, more specific testing is performed to determine which particular antibody profile exists. The accuracy of the test appears to be higher in teenagers and adults when compared to young children. As the sensitivity for this testing is not high, many gastroenterologists utilize the ESR as a better, cheaper screening test.
Dr. Listernick: What are the CT findings of Crohn's disease?
Dr. Rigsby: We look for thickened, irregular bowel folds and abscesses adjacent to bowel walls, suggestive of transmural inflammation. Another CT scan finding is that of "creeping fat" encroaching on the bowel loops. While not specific for Crohn's disease, it is highly suggestive if the clinical details point to IBD.
Dr. Listernick: For a pediatrician who is investigating a child for Crohn's disease, what is the best initial radiographic study to perform?
Dr. Rigsby: That probably depends at which institution you trained and the skill of your radiologists. Although a barium study of the small intestines gives a better look at the mucosal pattern, the CT scan will give you a picture of the wall, creeping fat, intraabdominal abscesses, and the colon.
Dr. Listernick: What were the findings on this child's most recent scan?
Dr. Rigsby: The lymphadenopathy is even larger, displacing bowel loops. There were several new findings, including extensive mesenteric and marked thickening of both the distal small bowel loops and the entire colon.
Dr. Listernick: Given the clinical and the radiographic findings, she underwent colonoscopy.
Dr. Li: Despite the findings on CT scan, the endoscopic findings were unimpressive. The colonic mucosa appeared to be entirely normal. On biopsy, we found mild ileal inflammation, but no specific pathologic correlates of IBD and not enough inflammation to account for an ESR of 100 mm per hour.
Dr. Listernick: We don't have a diagnosis. What do we do now?
Morris Kletzel, MD, pediatric oncologist: Give me a needle and I'll make the diagnosis. While I can't give an extensive differential diagnosis at this point, we clearly need to examine the lymphadenopathy and the bowel wall. Malignancy would be high on my list based on the current CT scan findings. I recognize that there aren't a lot of malignancies that would produce symptoms on and off for 1 8 months without "coming to a head" sooner. One that comes to mind is Rosai-Dorfman disease (sinus histiocytosis). The clinical picture includes massive lymphadenopathy, fever, leukocytosis, and a high ESR. The lymph node histology reveals a dense infiltrate of pathognomonic histiocytic or macrophage-like cells that have characteristic immunohistochemical features. This is often a selflimited disease that has a waxing and waning course. Treatment may be required when the massive lymphadenopamy interferes with adjacent organ function.
Dn Listernick: This child's clinical presentation was very confusing, particularly the evanescent episodes of inflammatory symptoms and pain during such a long period. Is an exploratory laparotomy reasonable at this point?
Robert Arensman, MD, pediatric surgeon: Absolutely, given the clinical course. Laparoscopy, followed by laparotomy if necessitated by me findings, would allow for a minimally invasive procedure that would allow for lymph node and mesenteric biopsies.
David Hoover, MD, pediatric surgeon: At the time of surgery, we first discovered bloody peritoneal fluid, which we generally only find in abdominal trauma or malignancies. There was a dense fibrotic mass at the base of the mesentery, adjacent to the bowel. This mass extended throughout the retroperitoneum, preventing me bowel from being freely mobile. We were able to perform an incisional biopsy of the mass. To remove it in its entirety would have necessitated a radical dissection and extensive resection of the bowel itself.
Dr. Listernick: You say that the mass is retroperitoneal. Why don't we see that on the CT scan?
Dr. Rigsby: In retrospect, I believe that the mass probably consists of the lymphadenopathy and the inflammatory reaction that we see on the CT scan, which is at the base of the mesentery just at the beginning of the retroperitoneum.
Dr. Shulman: Before we see the pathology, we recently have seen several children who have had a fibrosing retroperitoneal process that led to ureteral obstruction and hydronephrosis as a presenting sign of chronic granulomatous disease.
Dr. Listernick: Flow cytometry for chronic granulomatous disease was negative in this girl.
Wei Hsueh, MD, pediatric pathologist: The mass was composed of loose fibrous tissue mat was infiltrating the surrounding tissues in an aggressive fashion. However, the cytology of the cells appeared benign. In addition, there were a lot of inflammatory cells within the tissue, including eosinophils, lymphocytes, and plasma cells. Special staining for muscle cells revealed that the fibrous tissue was comprised of myofibroblasts. This mass was an inflammatory myofibroblastic tumor (IMT), previously known as either inflammatory pseudotumor or plasma cell granuloma.
Dr. Listernick: Is this a neoplasm?
Dr. Hsueh: It is definitely a neoplasm. There have been several studies that have identified the clonal nature of this tumor. The new World Health Organization tumor classification labels this an intermediate-grade tumor. This tumor often presents as a solitary mass; common sites include the orbit, bladder, lung, liver, kidney, and mesentery. IMT rarely metastasizes; however, large tumors that have spread diffusely, as in this case, may have a worse prognosis due to their infiltrative nature.
Dr. Listernick: What is the pathogenesis of the inflammatory symptoms?
Joanna Weinstein, MD, pediatric oncologist: IMT has been shown to release a variety of cytokines that cause these symptoms. Although the pathogenesis of the tumor is unclear, it has been linked to a number of inciting events including trauma and infection.
Dr. Listernick: What is the treatment for IMT?
Dr. Weinstein: Removal of the tumor is generally curative; the inflammatory symptoms disappear. Diffuse tumors such as occurred in this child are more problematic; there are reports of extensive surgeries with radical resections of bowel, often associated with a poor prognosis. Corticosteroids, conventional chemotherapy, and radiation therapy all have had mixed results. Several groups have identified angiogenic factors such as vascular endothelial growth factor and fibroblastic growth factor in these tumors. This has led researchers to look at the antiangiogenic properties of cyclooxygenase-2 inhibitors as a potential treatment. A recent report documented the shrinkage of a tumor similar to this patient's associated with the use of anti-inflammatory medication. We elected to treat her with naproxen and will follow her clinically as well as with serial abdominal CT scans.
Dr. Listernick: Can we follow the tumor with magnetic resonance imaging (MRI) scans so that we don't expose her to so much ionizing radiation?
Dr. Rigsby: Unfortunately, MRI does not visualize the mesentery and the bowel well. I would use CT scans if you want to get the best images.
Dr. Listernick: Thank you, everybody.