Tuberculous Mesenteric Lymphadenitis Presenting as a Pancreatic Mass

An unusual presentation of tuberculosis can be a difficult diagnostic dilemma and can lead to delays in making a correct diagnosis and initiation of appropriate therapy. A negative culture can further delay initiation of effective treatment in drug-resistant tuberculosis. CASE REPORT A 28-year-old female from Ulaanbaatar, Mongolia, arrived in the United States on February 23, 2000, to serve as a missionary for her church. As part of her visa requirement, she had a chest x-ray in late August 1999, which was read as normal. She spent 3 weeks in Provo, Utah, and then was assigned to a mission in Charlotte, North Carolina. On July 3, 2000, she presented at an emergency room complaining of abdominal pain of about 6-hour duration. The pain and associated tenderness were located in the epigastric and periumbilical areas on presentation but localized in the right lower quadrant overnight. Initial laboratory testing revealed a white blood cell count of 13,500. An upper gastrointestinal endoscopy was normal. The initial abdominal computed tomography (CT) scan without contrast on July 3 was read as normal, but a subsequent CT scan with contrast the next day showed inflammatory changes around the medial portion of the cecum. She was clinically diagnosed with acute appendicitis and was taken to surgery on July 4, 2000, where an acutely inflamed appendix was removed. Pathological examination revealed a suppurative and gangrenous appendix without evidence for rupture or leak. The operative report states that manual and visual inspection of the rest of the abdomen was unremarkable. She did well postoperatively and returned to her missionary activities. Several months after this procedure, she began to experience right upper quadrant pain, which she described as sharp, sometimes as burning, and at other times as a pulling sensation with radiation around her side and into her back. She stated that food generally decreased the pain. She had no associated nausea, vomiting, or weight loss. She stated that the pain could last 10 minutes or all day and frequently caused her to double over which would give partial relief. On November 21, 2000, she underwent an abdominal ultrasound, which was read as normal and included a comment that the pancreas was within normal limits. She next presented at a hospital on July 6, 2001, stating that she had had continuous pain for approximately 3 months. The pain was localized to the midepigastric and right upper quadrant areas, and she noted it to be significantly tender in those areas. There was a palpable 3 to 4-cm mass in the right upper quadrant towards the midline which was somewhat tender. She had no rebound or guarding. On July 13, 2001, she underwent an abdominal CT scan with contrast, which revealed a cystic mass which measured 2 × 5 × 2 cm located in the gastrohepatic ligament adjacent to the superior aspect of the neck and head of the pancreas. At that time, she was transferred to Salt Lake City for further care. After arriving in Salt Lake City on August 6, 2001, she was seen by a general surgeon, who evaluated and confirmed the previous findings. A repeat CT scan revealed the multicystic mass measuring 5.6 cm transversely by 3.4 cm anterior-posteriorly (Fig. 1). The lesion was located immediately superior to the head and neck of the pancreas and medial to the portal vein. The majority of the body, tail, and uncinate process of the pancreas and inferior portion of the head of the pancreas were normal. There was no evidence of pancreatic ductal dilatation. At that time, a CT-guided biopsy of the peripancreatic mass was undertaken, and 3 specimens were obtained through a 20-gauge needle. Pathology sections revealed mesothelial and inflammatory cells and inflamed fat. There were no malignant cells identified. The patient continued to have increasing pain and therefore was scheduled to have abdominal exploratory surgery, which was accomplished on September 17, 2001. At surgery, there were several enlarged nodes palpable within the gastrohepatic ligament, which were removed and sent for frozen section. Results showed chronic inflammatory tissue with no evidence of malignancy. Visualization of the pancreas revealed 2 large palpable cystic structures adjacent to it. They were biopsied, and the smaller of the 2 was excised. The larger one was not removed because of its close proximity to the gastroduodenal artery, common hepatic artery, as well as the portal vein. The pancreas itself appeared to be normal. Multiple sections of the specimen submitted revealed both soft tissue and lymph nodes extensively involved by necrotizing granulomatous inflammation. The granulomas were of different sizes and shapes and different age. Silver and acid-fast stains did not reveal any microorganisms. Microbiologic studies revealed a few morphologically typical mycobacteria on smears of lymph nodes. The pancreatic biopsies were normal. Other workup at that time included a normal chest x-ray, 3 negative sputum smears for acid-fast bacilli, a negative human immunodeficiency virus (HIV) test, and a purified protein derivative test with 14 mm of induration and some associated blistering as well. A clinical diagnosis of mesenteric lymphadenitis due to Mycobacterium tuberculosis was made, and she was started on 4-drug therapy including isoniazid, rifampin, pyrazinamide, and ethambutol, which was to continue for 2 months. She continued to do well postoperatively. However, at 8 weeks, all acid-fast bacilli cultures remained negative. After completing 8 weeks of therapy, the ethambutol and pyrazinamide were stopped, and isoniazid and rifampin were continued. Shortly thereafter, she returned to her home in Mongolia. She completed approximately 1 year of therapy with antituberculous medications. Sometime near the end of the year, she began to complain of swollen lymph nodes in her neck, which progressed to open ulcers with drainage of purulent material from her cervical lymph nodes bilaterally. She was seen by a physician in Mongolia who took her to surgery on 2 different occasions to excise infected tissue, but no cultures were available to identify the cause of the infection. After several relapses of drainage and increasing pain, we were able to get them to collect specimens aseptically from these draining sinuses in her neck and send to our laboratory in Salt Lake City for culture. The smears were positive for acid-fast organisms again and on culture grew Mycobacterium tuberculosis, which was identified and confirmed by deoxyribonucleic acid sequencing. Susceptibility testing revealed the organism to be resistant to isoniazid, rifampin, ethionamide, pyrazinamide, and streptomycin. It was susceptible to ofloxacin, cycloserine, capreomycin, ethambutol, and paraaminosalicylic acid.FIGURE 1: CT scan demonstrating a multicystic mass superior to the head of the pancreas.We were able to arrange shipment of medications for her to Mongolia to be administered by the mission physician located there. After consultation with physicians at the National Jewish Hospital, she was started on levofloxacin, cycloserine, and paraaminosalicylic acid granules. She is currently on this 3-drug regimen, and she is slowly improving. DISCUSSION Although the incidence of extrapulmonary tuberculosis has not decreased as rapidly as pulmonary tuberculosis, extrapulmonary disease remains an uncommon manifestation of tuberculosis in the United States in the absence of HIV infection. Mesenteric lymphadenitis due to tuberculosis is even less common.1 Extrapulmonary tuberculosis can occur in almost any body site or organ, but pleural, skeletal, central nervous system, genitourinary, miliary, and gastrointestinal sites are particularly prevalent. The pathogenesis of extrapulmonary tuberculosis occurs by 3 mechanisms: (1) superficial mucosal foci due to the spread of infectious pulmonary secretions via the respiratory and gastrointestinal tracts, (2) foci established by contiguous spread, and (3) foci established by lymphohematogenous dissemination, either at the time of primary infection or less commonly from chronic established pulmonary or extrapulmonary foci. In the absence of immunosuppression, the latter is very uncommon.2 Before the advent of effective chemotherapy for tuberculosis, over 70% of patients with advanced pulmonary disease developed gastrointestinal disease from swallowing infected secretions. Disease could be located anywhere in the gut including the esophagus, stomach, and the small and large intestines and could involve any gastrointestinal-associated organ, that is, spleen, liver, and pancreas as well. In a series of 81 patients with abdominal tuberculosis reported from Canada by Jakubowski et al,3 41 patients had peritoneal disease, 17 had ileocecal involvement, and 16 had anorectal disease. Eight patients had mesenteric lymphadenitis, and one patient each had hepatitis and granulomatous disease in the sigmoid colon. In the absence of evidence of active pulmonary disease, the mesenteric lymphadenitis seen in our patient is probably a result of lymphohematogenous dissemination at the time of the primary infection. When the primary infection occurred is not clear. However, mesenteric adenitis is thought to represent an early form of tuberculosis,3 is seen predominately in Asians and young women, and usually occurs in individuals from countries where the prevalence of tuberculosis is high. All these factors are consistent with the reported patient. Abdominal masses and pain constitute a common presentation, and concern that cancer is the cause is frequently an issue. A significant delay from the time of immigration to presentation with symptoms is also common. In a series of 22 immigrants to London with peritoneal tuberculosis from Africa and the Indian subcontinent, the time from immigration to presentation with symptoms was reported. The mean delay was 5.2 years (range 1 to 12 years), suggesting that the patient's vulnerability to this disease is brought into the new country and persists for many years. The trigger for activation remains unknown.4 Our patient was treated with 4-drug therapy for 8 weeks while awaiting culture results. When the cultures remained negative, we elected to continue treatment with isoniazid and rifampin alone since she was returning to Mongolia where little or no follow-up was possible. This decision seemed rational at the time but proved not to be a good one. After several months of therapy, she developed open draining sinuses bilaterally from the cervical lymph nodes. Successful culture of this material confirmed our suspicions that she had multidrug-resistant Mycobacterium tuberculosis. This site was probably seeded by lymphohematogenous spread at the time of the initial infection or possibly later during ineffective chemotherapy of her mesenteric lymphadenitis. A mini-chest x-ray taken at this time and transmitted electronically from Mongolia was reviewed by a radiologist and interpreted as negative. Two types of intraabdominal tuberculosis associated with the pancreas have been reported with significant overlap. They may be the same process depending on the original site of infection and the extent of invasion by the bacterium. Both present with abdominal pain, fever, and often a palpable mass. Both are frequently confused clinically with pancreatic cancer. The first is actual pancreatic involvement with tuberculosis and usually will require surgical drainage of an intrapancreatic abscess both for diagnosis and treatment.5-8 The second is tuberculous involvement of the mesenteric lymph nodes in close proximity to the pancreas. Frequently, the CT scan cannot differentiate the mass from the pancreas. A magnetic resonance scan may be more sensitive for this purpose.9,10 Our patient fit into the second category, although the CT scan did localize the multicystic mass as immediately superior to the head and neck of the pancreas. Under direct visualization at surgery, the pancreas did not seem to be involved, and the biopsies of the pancreas confirmed this finding. Published reports emphasize that most patients with pancreatic tuberculosis and mesenteric lymphadenitis occur in persons from developing nations especially from Asia where the prevalence of tuberculosis is high. Mesenteric lymphadenitis affects mainly young Asian women in the third and fourth decades of life consistent with our 28-year-old female patient from Mongolia. Preoperative needle biopsy of the mass in our case did not establish a diagnosis. Even with better imaging studies, the differentiation between pancreatic cancer and tuberculous mesenteric lymphadenitis can be difficult, and most patients will require laparotomy to establish the diagnosis. After initiation of 3-drug therapy to which the organism was susceptible, the draining sinuses closed and healed, and the patient's sense of general well being has rapidly improved. The treatment plan is to complete 24 months of triple-drug therapy.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.